‘State threats’, security, and democracy: the National Security Act 2023

Abstract

The National Security Act 2023 replaces the Official Secrets Acts 1911, 1920 and 1939, updating, rationalising, and expanding the various offences which they contained and introducing new rules aimed at the same broad end of countering the threat posed to the UK by the efforts of hostile states and their proxies. It therefore represents a legislative confirmation of the ongoing pivot back to ‘state threats’ rather than terrorism as the focus of the national security enterprise in the UK, though now informed by the experience of counter-terrorism law since 2000. This paper assesses the main changes made by the 2023 Act, including in the context of threats to the democratic process, actual and potential, which have been identified in recent years. The argument offered is that the focus of the 2023 Act – encompassing threats to democracy only where they rise to the level of threats to national security – is undermined by the absence of a more thoroughgoing project to protect the democratic process more generally against foreign interference.

‘The intelligence community have not notified us of any successful attempts to interfere in UK elections. As I mentioned, the Electoral Commission is not a national security body – we do not have intelligence functions – so when it comes those matters, we receive the information rather than creating it or analysing exactly what it means.’¹

Introduction

Modern national security law in the UK has developed in a series of cycles, prompted by the changing legal landscapes and the (re)emergence of new and different threats to the state and its interests. One such cycle began in the 1980s, as the influence of the European Convention on Human Rights (ECHR) forced the UK to give to the security and intelligences agencies (SIAs) and to practices of interception a more solid legal base than they had previously enjoyed.² Another took place around the year 2000, when the law of counter-terrorism entered an era in which terrorism was recognised as a global phenomenon rather than one associated largely with Northern Ireland, though the new regime put in place by the Terrorism Act 2000 was quickly overtaken by the post-September 11 impulse to constantly rethink the law in response to every new trend or threat.³ The most recent such cycle has now taken in a body of law that was last addressed during the Cold War: the law of official secrets, largely unamended since 1989.⁴ The National Security Act 2023 (NSA 2023 or the 2023 Act) replaces the Official Secrets Acts 1911, 1920 and 1939 – but not, importantly, that of 1989 – updating, rationalising, and expanding the various offences which they contained and introducing new rules aimed at the same broad end of countering the threat posed to the UK by the efforts of hostile states and their proxies. It therefore represents a legislative confirmation of the ongoing pivot back to ‘state threats’ rather than terrorism as the focus of the national security enterprise in the UK – of a sort evidenced, most strikingly, by the work of the Intelligence and Security Committee of Parliament in recent years⁵ – though now informed by the experience of counter-terrorism law since 2000.

This paper assesses the main changes made by the 2023 Act. The first section considers the changes it does and does not make to the law of official secrets, against the background of the detailed work done on that body of law by the Law Commission.⁶ The second part considers those aspects of the Act aimed at ‘state threats’ which exist beyond the criminal law. It provides, first, for a new civil order, backed by criminal penalties, which is modelled on the various administrative schemes introduced within counter-terrorism law in recent decades. Alongside that is a ‘foreign influence registration scheme’ (FIRS), which seeks to capture the work of non-state actors on behalf of foreign states, but which is significantly narrower than that which the Government originally proposed. These changes, though welcome insofar as they end the modern focus on counter-terrorism to the exclusion of almost everything else, bring back into play bigger questions – mostly absent in the era of national security as largely, if not exclusively, counter-terrorism – about the relationship between national security law and democracy. The final section of the paper therefore considers the Act in the context of threats to the democratic process, actual and potential, which have been identified in recent years. The argument offered is that the restricted focus of the 2023 Act – encompassing threats to democracy only where they rise to the level of threats to national security – would seem logically to assume the existence of an underlying project to protect the democratic process more generally against foreign interference: a project of a sort which has not so far been pursued with any vigour in the UK. In the absence of such a project, and in particular interventions made to exclude foreign money from the British political process, the Act leaves the door open to threats to democracy which are perhaps more mundane but nevertheless more pernicious than are those against which it directly guards.

1. The turn (back) to state threats

The presence in the short title of the 2023 Act of the term ‘national security’ is itself notable, for though that language has supplanted the older term ‘defence of the realm’ (largely, it would seem, under the influence of the ECHR)⁷ it has only in recent years found its way into the short title of statutes in the UK.⁸ But it is of course a broad formulation. It encompasses, for present purposes, two matters which are distinct, though which sometimes overlap in practice. One is counter-terrorism, which has been the focus in the UK since the events of 2001 – being the subject of a large number of legislative interventions, notwithstanding that the matter had been dealt with in an apparently exhaustive fashion by the Terrorism Act 2000. But in longer perspective, taking in much of the history of the relevant security bodies, this is the exception rather than the rule. Until the end of the Cold War, the security and intelligence agencies focused instead upon the question of the other aspect of national security, which we might call ‘state threats’ – the formulation which was originally intended to be used in the title of the 2023 Act.⁹ In recent years, following the collapse of the ‘Islamic State’, there have been a number of indications that state threats – emanating primarily, it would seem, from Russia and China, but also Iran¹⁰ – have become once again central to the national security enterprise in the UK. Certainly, this appears to be the understanding of the SIAs themselves. When the Director General of MI5 gave its annual threat update in 2021, for example, he began with the issue of state threats, only thereafter turning to what he acknowledged was ‘still the national security threat of greatest concern to the public’: terrorism.¹¹ It was necessary, he said, to ‘build the same public awareness and resilience to state threats that we have done over the years on terrorism’.¹² One aspect of that project was the enactment of what has in time become the National Security Act, of which not only the current Director General but also a number of his predecessors have been vocal supporters, including in Parliament.¹³

In fact, though the NSA 2023 is now the centrepiece of the legislative response to state threats, steps had already been taken in this direction, in the form of powers introduced by the Counter-Terrorism and Border Security Act 2019,¹⁴ modelled on those under Schedule 7 to the Terrorism Act 2000.¹⁵ Where the Schedule 7 powers are available in order to determine whether a person appears to be involved in terrorism, those in the 2019 Act permit the questioning of persons at ports ‘for the purpose of determining whether the person appears to be a person who is, or has been, engaged in hostile activity’.¹⁶ A person, the statute provides, is engaged in hostile activity if he or she ‘is or has been concerned in the commission, preparation or instigation of a hostile act’ which is or may be ‘carried out for, or on behalf of, a State other than the United Kingdom’ or is ‘otherwise in the interests of a State other than the United Kingdom’.¹⁷ An act is hostile if it ‘threatens national security’ or ‘the economic well-being of the United Kingdom in a way relevant to the interests of national security’ or is ‘an act of serious crime’.¹⁸ The key feature of this power is that it can be exercised even where are no grounds for suspecting that the person is or has been engaged in such activity.¹⁹ Amongst the consequences of its use are that a number of ancillary powers are brought into play. These include, for example, the power to search the person being questioned and retain, on one of a number of grounds, articles found there; copies may also be made, and those copies may be retained on equivalent grounds. A person questioned under these powers must give over any information in his or her possession that is requested by the examining officer.²⁰ One important effect is that, as with Schedule 7,²¹ these powers can be used to bypass the process which would ordinarily have to be followed in order to require the provision of a password under the Regulation of Investigatory Powers Act 2000.²²

2. The new law of espionage

A proposal to update the law on unauthorised disclosure found in the Official Secrets Act 1989 having been set aside – as a result, it would seem, of disagreement on the desirability of a public interest defence²³ – the reform of criminal law by the National Security Act is more modest than was first intended. It replaces, that is, only the older parts of the law of official secrets: the offences found in the Official Secrets Acts 1911, as augmented and amended by Acts of 1920 and 1939, but addresses a much broader range of threats than do those statutes – not only the traditional espionage (theft of secret information, for example) at which the old law had aimed but also various forms of political or electoral interference. In its work on the topic, the Law Commission had recommended a new, modernised, law of espionage in place of these older offences, with which a number of issues were identified.²⁴ Those subsequently put in place by the 2023 Act have a number of key features. Many of them have as a constitutive element the ‘foreign power condition’ (FPC) whose definition is therefore central to the scope of those offences as well as other, civil law, elements of the Act into which it feeds. The condition is met in relation to a person's conduct where the conduct, or a course of conduct of which it forms part, is carried out ‘for or on behalf of a foreign power’ and the person either knows or ‘having regard to other matters known to them ought reasonably to know’ that this is the case.²⁵ A non-exhaustive set of examples of what it means for conduct to be carried out ‘for or on behalf of’ a foreign power includes where it is: ‘instigated by a foreign power’, ‘under the direction or control of a foreign power’, ‘carried out with financial or other assistance provided by a foreign power for that purpose’ or ‘carried out in collaboration with, or with the agreement of, a foreign power’.²⁶ It can apply to conduct carried out directly on behalf of a foreign power, as well as that where the relationship with the foreign power is mediated, for example, by one or more companies.²⁷ It is also met, however, if a person merely intends the conduct to benefit a foreign power,²⁸ and so there is no requirement that there exist an understanding, or have been any contact, between the person and the foreign power in question.

A foreign power for these purposes is one of the following: ‘the sovereign or other head of a foreign State in their public capacity’, ‘a foreign government, or part of a foreign government’, ‘an agency or authority of a foreign government, or of part of a foreign government’, ‘an authority responsible for administering the affairs of an area within a foreign country or territory, or persons exercising the functions of such an authority’ or ‘a political party which is a governing political party of a foreign government’.²⁹ It would therefore include, for example, not only the Government of the Russian Federation, say, but also the Chinese Communist Party or its ‘United Front Work Department’.³⁰ No attempt is made to exclude friendly or allied states from the definition,³¹ but political parties registered in the UK are excluded from the final category, as are parties in government in the Republic of Ireland.³² Apart from that, however, there is no limitation on which states are captured by the definition: ‘foreign government’ means ‘the government of a foreign country or territory’ and ‘foreign country or territory’ means ‘a country or territory outside the United Kingdom, the Channel Islands, the Isle of Man or the British Overseas Territories’.³³ All foreign countries, as well as the constituent territories of federal states, are therefore foreign powers for these purposes.

The effect of the FPC is that conduct which is otherwise lawful, constitutes only a more minor offence, or is wrongful only as a matter of private law, is an offence if carried out on behalf of, or with the intention of benefiting, some other (part of a) state or its government. The offences to which it applies include, first of all, an obvious analogue of one found in the 1911 Act: that of obtaining, copying, recording, retaining, or disclosing or providing access to, protected information, for a purpose that the person knows or ought to know is ‘prejudicial to the safety or interests of the United Kingdom’.³⁴ Protected information is ‘any information, document or other article’ where, for the purpose of protecting that safety or those interests, access to it is restricted where, or it is reasonable to expect access to be so restricted.³⁵ This formulation – ‘the safety or interests of the United Kingdom’ – recurs throughout, and the Government has been clear that its intention is that the meaning given to it will be that given, in Chandler v DPP,³⁶ to the equivalent language in the OSA 1911: not, that is, the interests of what Lord Pearce called the ‘amorphous populace’ but rather ‘the interests of the State according to the policies laid down for it by its recognised organs of government’.³⁷ A second offence, not defined by reference to the ‘safety or interests of the United Kingdom’, but to which the FPC also applies, forbids obtaining or disclosing trade secrets without authorisation.³⁸ There is also an offence of sabotage, where a person intentionally or recklessly engages in conduct which results in damage to an ‘asset’ for a purpose prejudicial to the safety or interests of the UK.³⁹

There are then two offences, without equivalent in the previous law, of ‘foreign interference’ – one encompassing interference in elections, the other general. The former applies where a person commits a ‘relevant electoral offence’ and the FPC is met. The offences in question include classic electoral offences such as personation and treating, but also offences relating to making and declaring political donations.⁴⁰ No otherwise lawful conduct is caught by this new offence; rather, a new offence, with harsher penalties, applies where the underlying offence is committed in circumstances in which the FPC applies. The general foreign interference offence is structured in a complex fashion, applying where (the FPC being met) a person engages in conduct with the intention that it has one of the number of effects, including interfering with ‘the exercise by a particular person of a Convention right in the United Kingdom’, ‘the exercise by any person of their public functions’, ‘whether, or how, any person makes use of services provided in the exercise of public functions’, ‘whether, or how, any person… participates in political processes under the law of the United Kingdom’ and ‘whether, or how, any person… participates in legal processes under the law of the United Kingdom’.⁴¹ One of three conditions must also be met for conduct to be an offence under this provision: the conduct in question must otherwise be an offence, involve coercion of any kind (including, for example, threats to damage a person's reputation, or the imposition of undue spiritual pressure), or involve the making of a misrepresentation.⁴² This offence therefore captures both acts aimed at private individuals and at public actors. It might capture, for example, attempts (at the behest of a foreign power) to threaten a private person in order that he or she not give evidence in court, to coerce a person into voting a certain way in an election, or to blackmail an MP into expressing (or not) certain views.

There are other offences to which the FPC does not apply because the activities in question are inherently supportive of foreign powers. These include intentionally assisting a foreign intelligence service in carrying out UK-related activities or engaging in conduct ‘that is likely to materially assist a foreign intelligence service in carrying out UK-related activities’ where the person ‘knows, or having regard to other matters known to them ought reasonably to know, that their conduct is likely to materially assist a foreign intelligence service in carrying out UK-related activities’.⁴³ UK-related activities are those which take place in the UK or take place elsewhere but which are prejudicial to the safety or interests of the UK.⁴⁴ It is also an offence to obtain, or agree to obtain, a material benefit from a foreign intelligence service (directly or indirectly) where the person knows or ought reasonably to know that the benefit in question is being provided by such a service.⁴⁵

A range of associated provision is made, perhaps most notable amongst it being the power – analogous to one found in the OSA 1920⁴⁶ – to exclude the public from any part of proceedings (other than the passing of sentence) for an offence under Part 1 of the Act.⁴⁷ The Act also introduces a number of powers of investigation in relation to this suite of offences: not only traditional powers of entry, search and seizure, but also, for example, ‘customer information orders’⁴⁸ (allowing authorities to require financial institutions to hand over any information they have relating to a specified person) and ‘account monitoring orders’ (allowing authorities to require such institutions to hand over specified information about accounts held there).⁴⁹ Many of these powers, along with a power of warrantless arrest, are framed by reference not directly to specific offences under Part 1 of the 2023 Act, but rather to the concept of ‘foreign power threat activity’, which also plays a significant role elsewhere in the legislation. That concept is defined in turn by reference not only to the acts which constitute a number of (but not all of) the offences discussed above but also to acts (or threats to carry out acts) which ‘involve serious violence against another person’, ‘endanger the life of another person’ or ‘create a serious risk to the health or safety of the public or a section of the public’, but only where the FPC is met.⁵⁰ It includes not only the ‘commission, preparation or instigation’ of the relevant acts or threats, but also ‘conduct which facilitates (or is intended to facilitate)’ conduct in that category, as well as conduct which involves knowingly giving support or assistance to someone carrying out such conduct.⁵¹ The effect is to create a broad category of activity which brings into play highly-intrusive investigative capabilities, so that the state will be able to monitor in detail the financial activity of those suspected of involvement, even quite indirectly, in a range of activity by foreign states aimed at the interests of the UK.

3. Prevention and investigation measures

Though the reform of the Official Secrets Acts was the initial impetus for the 2023 Act, the statute enlists also civil law in its attempts to counter the problem of ‘state threats’. One way in which it does so is by introducing a new civil order which might be imposed on persons so as to ‘mitigate the risk posed by individuals engaged in hostile activity’.⁵² As with the power of suspicionless stop and search introduced by the Counter-Terrorism and Border Security Act 2019, this new mechanism, the State Threats Prevention and Investigation Measure (STPIM),⁵³ has a clear precursor within the law of counter-terrorism, being modelled – explicitly and unambiguously – on the Terrorism Prevention and Investigation Measures (TPIMs) provided for by the Terrorism Prevention and Investigation Measures Act 2011 and modified at frequent intervals thereafter.⁵⁴ STPIMs – intended, it was said, to be a ‘tool of last resort’ where criminal prosecution is not possible⁵⁵ – are framed by the same concept of ‘foreign power threat activity’ discussed above, reasonable belief of an individual's involvement in which is the key precondition for the imposition of a STPIM.⁵⁶ Imposition of an STPIM requires, except in cases of urgency, the prior permission of a court, the role of which is, however, limited to considering whether the relevant decisions of the Secretary of State that the conditions for the imposition of a PIM are satisfied are ‘obviously flawed’.⁵⁷ And in determining that the court must apply ‘the principles applicable on an application for judicial review’.⁵⁸ All of this is familiar from the TPIM context.

And again, as with TPIMs (amendments to which have made them more potentially intrusive than was originally the case), the measures that can be imposed on an individual suspected of involvement in foreign power threat activity are broad and potentially highly intrusive.⁵⁹ They include: measures relating to a person's residence (requiring a person to reside at a specific location, for example, or to remain at a residence between specified hours); relating to travel (to not leave the UK, for example, or to surrender a passport); relating to a person's use of financial services (to hold only one bank account, for example, and to close any others one holds); relating to the transfer or ownership of property; relating to the use of electronic communication devices; relating to a person's association or communication with others; and a variety of measures relating to monitoring compliance with any measures imposed.⁶⁰ An STPIM notice lasts for one year, but can be extended up to four times if the relevant conditions are met,⁶¹ giving a maximum duration of five years in total. The investigative element of the STPIM regime, whereby the Secretary of State is required to consult police as to whether there is evidence that might be used for a prosecution for one of the acts constituting foreign power threat activity and the police are under a duty to secure that the investigation of the individual is kept under review while a notice is in force, reflects the same logic which was intended to distinguish TPIMs from its predecessors, and in particular ‘control orders’. Whether this will amount to much in practice is doubtful: the TPIM regime appears not to have been the catalyst for major changes in the investigation and prosecution of terrorism offences. How significant these new powers are will depend on the frequency with which they are used, but it is worth noting that reliance on the TPIM regime has faded over time, with only one TPIM in force as of mid-2023.⁶²

But even if the investigative element of STPIMs proves to be largely symbolic, because the imposition of a measure does little to increase the likelihood of a prosecution for conduct relevant to foreign power threat activity, that does not exhaust the role of the criminal law here. We noted above the power to exclude the public from proceedings for one of the new espionage offences. But an in camera procedure does nothing to limit the availability of the sensitive material to the person being prosecuted. The procedure which does permit a reliance on such material without it being disclosed to the non-state party – the closed material procedure – is unavailable in (and only in) criminal proceedings.⁶³ And so material that is too sensitive to reveal not only to the public but also to the person being charged – such as, perhaps, information about sources of intelligence, human or otherwise – cannot be directly used to support a criminal prosecution under this Act or any other. The caveat ‘directly’ is though crucial, for material that would not ordinarily be admissible in criminal proceedings, including material which is the product of the interception of communications and which is rendered inadmissible in legal proceedings generally by the Investigatory Powers Act 2016,⁶⁴ may be used within a closed procedure so as to justify the imposition of an STPIM,⁶⁵ and yet not disclosed to its subject (unless compliance with Article 6 requires the disclosure of the material or a ‘gist’ of it).⁶⁶ And because, in turn, a breach of the restrictions imposed under an STPIM without reasonable excuse (unless the restriction is a travel measure and the person leaves the UK, in which case there is no defence of reasonable excuse) is a criminal offence,⁶⁷ it is then possible to prosecute a breach of those restrictions in open court without making – or needing to make – reference to the material which originally justified it. This is a standard feature of the modern counter-terrorism landscape, and it is hardly surprising to see it extended to include the state threats domain. No longer will the need to keep secret sources and methods act as a total obstacle to the imposition of criminal sanctions on those spying for foreign powers.

4. The foreign influence registration scheme

Given that the Part 1 offences discussed above replace and supplement long-standing criminal offences, and the Part 2 regime largely transplants into the sphere of ‘state threats’ the TPIM regime, the key innovation in the National Security Act is the ‘Foreign Activities and Foreign Influence Registration Scheme’.⁶⁸ This issue was not addressed in the Law Commission's work, and though it was argued for in the Government's response to the Commission report,⁶⁹ the scheme was not included in the Bill when it was first introduced into Parliament.⁷⁰ The background is suggested by the Intelligence and Security Committee's Russia report of 2020, which noted that ‘the arrival of Russian money resulted in a growth industry of enablers – individuals and organisations who manage and lobby for the Russian elite in the UK’:

Lawyers, accountants, estate agents and PR professionals have played a role, wittingly or unwittingly, in the extension of Russian influence which is often linked to promoting the nefarious interests of the Russian state.⁷¹

It observed that the Official Secrets Act did not address this issue, that ‘crucially, it is not illegal to be a foreign agent in this country’,⁷² and that schemes aimed at foreign influence operated elsewhere. Referring in particular to the scheme found in the United States Foreign Agents Registration Act, the ISC suggested that an equivalent would ‘clearly be valuable in countering Russian influence in the UK’.⁷³ In subsequently arguing for the scheme the Government identified two main benefits. One was that the existence of such a scheme would ‘provide a means of prosecuting known hostile actors without necessarily having to disclose the most sensitive evidence’ with the prosecution needing, for example, only to disclose ‘evidence of carrying out a registerable activity for or on behalf of a foreign state or foreign state actor’.⁷⁴ Second, a scheme of this sort could ‘provide a means to intervene at an earlier stage of the activity and before it results in a damaging hostile act’.⁷⁵ At the stage at which consultation on the proposal took place, a number of important questions remained open, and in the event the scheme as eventually enacted varies significantly from that first inserted into the Bill at Committee stage in the Commons, following significant concessions made when the Bill was in the House of Lords.

The effect of these changes was to significantly water down one of the two elements of the FIRS – what is described as the ‘primary’ or ‘political’ tier of the scheme. This tier targets the category of ‘foreign influence arrangements’,⁷⁶ in which a foreign power directs a person to carry out (or arrange to carry out) ‘political influence activities’ in the UK. Such an arrangement must be registered within 28 days; failure to register is an offence, as is carrying out registrable activity without having registered.⁷⁷ Political influence activities are defined broadly, so as to include any combination of a set of actions and a set of intentions. The intentions are those of influencing one of the following matters or persons: an election or referendum in the UK, a decision of central or devolved governments, the proceedings of a political party, or a member of the Westminster or devolved legislatures.⁷⁸ The activities which, if done with such an intention, give rise to a foreign influence arrangement include distributing money, goods or services to UK persons; making a public communication (‘except where it is reasonably clear from the communication that it is made by or at the direction of the foreign power’) or communicating with one of a range of persons – not just members of the UK's various executives and legislatures but also, for example, employees of MPs, members of the Senior Civil Service and senior military personnel.⁷⁹

When the Bill was before Parliament, however, attention was focused not for the most part on the activities covered by the primary tier of the FIRS, but rather the application of the scheme to the category of ‘foreign principals’, encompassing not just ‘foreign powers’ but also foreign companies and unincorporated organisations (other than Irish ones).⁸⁰ An arrangement by which a person agreed to lobby the government on behalf of a foreign NGO would, for example, have been a foreign influence arrangement, even if the NGO had no connection to the government of its home state. This breadth was widely criticised, often by those to whom the scheme as originally formulated would apply,⁸¹ and in particular for straying beyond the issue of national security into a more general regulation of lobbying.⁸² There was significant concern about unforeseen implications of the rules – whether they might catch, for example, British charities which received funding from foundations based overseas.⁸³ The decision to narrow the scheme – such that it applies only where the relevant activities are carried out at the behest of foreign powers (and not the broader category of foreign principals) – responds to some of those criticisms, but by definition leaves open a gap which hostile foreign states might exploit: a foreign unincorporated organisation which is aligned with a foreign power and staffed by its sympathisers but not directed by it can fund political influence activities in the UK without these being registered. Where it does apply, the effect of the scheme is not that such activity is forbidden but that it must be conducted openly, so that it might be judged accordingly. This is reflected also in the fact that the primary tier captures only arrangements with foreign powers, meaning that activity carried out by the foreign power itself is outside the scope of the FIRS, and its officials and employees may carry out political influence activities without registration so long as they do not misrepresent their activities or the capacity in which they act.

Alongside this ‘primary’ tier sits the ‘enhanced’ tier of the FIRS. Here, the Act permits the Secretary of State to specify in regulations a ‘foreign power’ (defined as in relation to the Part 1 offences, but with the exclusion of the Republic of Ireland) or another person (not an individual) controlled by such a power, where he or she considers it reasonably necessary to do so ‘to protect the safety or interests of the United Kingdom’.⁸⁴ Anyone who enters into a ‘foreign activity arrangement’ – one in which a specified person directs that person to carry out, or arrange to be carried out, activities in the UK – must register that arrangement within 10 days, it being an offence to fail to so register if one knows or (‘having regard to other matters known to them’) ought reasonably to know, that it is an arrangement encompassed by this requirement.⁸⁵ It is also an offence for a person to carry out activity pursuant to such an arrangement which is not registered.⁸⁶ What activities are covered by this requirement, however, depends on whether regulations have been made in relation to the specified person at issue: if so, then only those activities specified in the regulations are covered; otherwise all activities are within scope of the requirement. A specified person who is not a foreign power – such as a company owned or controlled by it – may not carry out the activities at issue in the UK unless those activities have been registered;⁸⁷ the same goes for its office-holders and employees,⁸⁸ who – in addition – are forbidden from carrying out activities in that capacity to the extent that they makes a misrepresentation about their activities or the capacity in which they act where the person is not registered.⁸⁹ As with the primary tier of the FIRS, the 2023 Act does not itself specify what information is to be provided by those to whom the enhanced tier applies, nor what will be done by way of publication or disclosure of information provided in accordance with the scheme, details of which will be left to later regulations. There can, though, be no requirement to disclose information in breach of legal professional privilege, or to disclose ‘confidential journalistic material’ or identify a ‘source of journalistic information’. It seems probable that the enhanced tier will be used sparingly – restricted, that is, to a small number of openly hostile countries and their proxies. Its effect will be to force those to whom it applies to choose between making their activities open and transparent or, more likely, continuing them notwithstanding that to do so without registration is now unlawful.

5. Democracy and national security

The 2023 Act aims – as its short title makes clear – specifically at threats to national security. It must be understood and evaluated, however, within the broader context of threats not to national security specifically, but rather to the democratic processes of the UK.⁹⁰ In recent years a number of concerns about such threats have been expressed, most famously by the ISC in its report on Russia. ‘Russian influence in the UK’, it said, ‘is “the new normal”, and there are a lot of Russians with very close links to Putin who are well integrated into the UK business and social scene, and accepted because of their wealth’.⁹¹ This may even in some cases have had the result of influencing the operation of the democratic process; of pushing the exercise of political power away from or towards certain objectives which suit the interests of Russia or some other foreign state: the ISC noted also that the UK is ‘clearly a target for Russia's disinformation campaigns and political influence operations’.⁹² Perhaps, as has been alleged of Russian involvement elsewhere, the point is simply to sow division,⁹³ making it unlikely that the country targeted will be able to present a united opposition to Russian interests. Perhaps, alternatively, the intervention seeks specifically to assist certain parties, or factions thereof, in gaining or retaining power. The ISC's report on China dealt again with the question of interference, including the difficulty of distinguishing it from legitimate attempts to influence foreign states.⁹⁴ It claimed that China ‘can be seen seeking to interfere with UK politicians, senior officials and military personnel, and they can be increasingly seen to interfere in the media, in Academia… and in relation to the Chinese diaspora’ but the section on electoral interference specifically was so heavily redacted as to obscure the issue of whether the UK is aware of specific attempts to interfere with electoral processes in the UK, of a sort that have been frequently alleged elsewhere.⁹⁵

Whatever the exact nature and intention of Russian and Chinese interference in the UK, so far only scattered efforts have been made to protect the integrity of the democratic process against this and similar threats. It is notable, therefore, that while the National Security Bill was before Parliament, the 2023 ‘refresh’ of the Integrated Review of Security, Defence, Development and Foreign Policy was published, in which it was noted that an area of vulnerability that had come into ‘sharper focus’ since the publication of the Review itself in 2021 was that of democratic resilience.⁹⁶ Alongside the Bill itself, the responses to that danger which were identified included not only the institution of the ‘Defending Democracy Taskforce’,⁹⁷ but also the forthcoming ‘Anti-Corruption Strategy’. So far little seems to have come of this work. The issue would appear to be vital: an ongoing case before the Special Immigration Appeals Commission relates to two Chinese nationals who were excluded from the UK because of their ‘involvement in providing financial donations to UK political figures on behalf of the Chinese Communist Party’.⁹⁸

Considered in this democratic context, we might observe that the framing of the 2023 Act – a legislative intervention targeted at, amongst other things, attempts by hostile state actors to subvert the democratic process illicitly – would appear to take for granted a background in which the democratic process is otherwise sound. The foundations, that is, are assumed to be solid, and need merely be protected against deliberate attempts to destabilise them. But there is reason for thinking that is not the case and, if and insofar as that remains true, the 2023 Act is liable not only to fail but also to mislead, giving the impression that the integrity of the democratic process is greater than it in fact is. Indeed, while the problems with the democratic process discussed in this section remain, the 2023 Act might have a perverse effect, incentivising those who wish to interfere with the UK's democratic processes to do so in a fashion which is more fundamental to those processes and so more consequential than are occasional attempts to influence specific individuals or Members of Parliament. Here, again, the Act must be understood within a longer development of the law in this area. The Conservative Party's 2019 manifesto included (at a time when the ISC's Russia report was still awaiting publication) a promise to introduce measures to ‘prevent any foreign interference in elections’.⁹⁹ What form of interference was left unspecified. The question of interference in the form of financial contributions, however, is a long-standing one, going back – at least – to the promise in the Labour Party manifesto at the 1997 general election that ‘Foreign funding will be banned’¹⁰⁰ and the subsequent recommendation of the Committee on Standards in Public Life that parties should ‘in principle be banned from receiving foreign donations’.¹⁰¹ This was the context of the rules, laid down in the Political Parties, Elections and Referendums Act 2000, whereby parties may accept donations only from ‘permissible donors’, defined so as to mean individuals on the electoral register and organisations of various stripes with adequate connection to the UK,¹⁰² with these rules backed up by offences relating to their evasion.¹⁰³

That these rules (and other restrictions put in place by the 2000 Act) may not suffice to exclude the use of foreign money to influence elections was considered in a 2021 report by the Committee on Standards in Public Life,¹⁰⁴ which highlighted a number of issues.¹⁰⁵ One was that rules on donations from companies provided ‘a potential route for foreign money into UK elections’.¹⁰⁶ The Committee therefore recommended that donations from companies should not be allowed to exceed net profits after tax generated in the UK over last two years.¹⁰⁷ Another was the possible use of unincorporated associations (UAs) to evade rules on permissible donors to political parties, in the absence of a requirement that the funds of such organisations themselves come from permissible donors and where there was no requirement ‘even to report (or, by implication, establish) full details of those who give them funds’.¹⁰⁸ The Committee concluded that ‘the rules on UAs are a weak point in the regime for regulating donations, and a potential route through which money from overseas sources can enter (and may already have entered) UK politics’,¹⁰⁹ and that they should therefore have to carry out permissibility checks for donations intended to fund political activity.¹¹⁰ In response, the Government said only that all of the recommendations regarding donations would be ‘considered carefully in detail to assess their viability in practice’.¹¹¹ Though it stated that it was considering issuing guidance which might ‘support campaigners to take a risk based approach to donations, similar to “know your customer/client” guidelines used in financial services, and undertake enhanced checks where appropriate’,¹¹² it later confirmed that it would not do so¹¹³ – an issue which was reopened in the context of the National Security Bill. The only recommendation of the Committee which the Government accepted, therefore, was that the Government should ban foreign organisations and individuals from buying campaign advertising.¹¹⁴ The relevant amendment was made by the Elections Act 2022,¹¹⁵ which places restrictions on campaign spending by third parties: those which are not, in effect, suitably closely linked to the UK.¹¹⁶ Such parties can only spend up to £700.¹¹⁷ But this limitation only applies during a regulated period – otherwise spending by these parties for political purposes is unlimited.

An amendment to the Elections Bill tabled by Liam Byrne would have gone further than this. Motivated by the belief, as it was put, that the Bill's worst flaw was ‘the lack of any attempt to clean up the laundromat of British politics, which is now awash with dark money from dubious sources’,¹¹⁸ it would have prevented a company donating more to a political party than the sum of its profits generated and taxable in the UK in the prior 12 months, and empowered the Electoral Commission to ‘call in’ a donation where it reasonably suspected that ‘a qualifying donation has given rise to or may give rise to a risk to national security in relation to electoral integrity’.¹¹⁹ It was not moved, but nevertheless called attention to the limits of the Government's ambitions in this regard and the possible routes for money to enter UK politics with a view to influencing attitudes or behaviours in certain areas or towards certain actors which persist even now.¹²⁰ Indeed, this problem is possibly exacerbated by another change made by the 2022 Act, which will in time permit overseas voters to remain on the electoral register – and so be a permissible donor to political parties¹²¹ – for life.¹²² The ability of a company operating in the UK to donate money not from UK profits remains intact, as does the ability of an unincorporated association suitably connected to the UK to donate money which came to it from outside the UK. And though parties are – as the Government frequently noted when the National Security Bill was before Parliament – obliged to check the eligibility of their donors, they are under no obligation to enquire as to the source of the money they are donating.

The NSA 2023, we see, does not purport to address the problem of foreign money in UK politics in general terms, but implicitly distinguishes between situations in which that money constitutes a threat to national security and those in which it does not. We have already considered above some of the key changes made: in particular the offences of general foreign interference and foreign interference in elections specifically. The latter will catch, for example, false declarations that the money being donated to a political party did not originate from some other source.¹²³ But the Act does not seek to obstruct in general terms the entry of foreign money into the UK political system. This fact was lamented by the Lord Evans of Weardale in the House of Lords,¹²⁴ whose views possess a heightened significance by virtue of his status as not only the current Chairman of the Committee on Standards in Public Life, but also a former Director General of MI5. ‘The proposals in the Bill’, he said, ‘are worthwhile, but they do not go far enough’:

They still leave a wide opportunity for, for instance, companies to donate into the electoral system even though they have not earned the money from which the donation would come in this country. Where has that money come from? It has come from abroad.¹²⁵

Reflecting these sorts of concerns, in the House of Lords an amendment – sponsored by, amongst others, Lord Carlile of Berriew, former Independent Reviewer of Terrorism Legislation – was made to the Bill, requiring political parties to publish, within three months of the Bill's passing, a policy statement ‘to ensure the identification of donations from a foreign power (whether made directly or through an intermediary)’ and thereafter to provide to the Electoral Commission an annual statement of risk management identifying ‘how risks relating to donations from a foreign power (whether made directly or through an intermediary) have been managed, and what measures have been put in place by the party to such effect’.¹²⁶ This reflects an effort, which the Electoral Commission has itself recommended, to introduce into the domain of political donations some of the concepts of anti-money laundering processes – a ‘know your donor’ requirement, as it has at times been put.¹²⁷

The amendment was, however, rejected by the House of Commons when the Lords amendments came to be considered there.¹²⁸ Though the Government took the view that the amendment was not needed,¹²⁹ the support which it had attracted in the Lords from those with national security experience – those who spoke in favour included not only Lord Evans and Lord Carlile, but also Lord West of Spithead (a member of the Intelligence and Security Committee), while the amendment was also supported by Lord Anderson (another former Independent Reviewer of Terrorism Legislation) and Lady Manningham-Buller (former Director General of MI5)¹³⁰ – was echoed by the Chair of the Intelligence and Security Committee, Julian Lewis. He suggested that the object of the amendment ‘should not be controversial’ and that it was ‘still not clear, despite the Minister's best efforts, why the Government would wish to oppose that clause’.¹³¹ The Government did though oppose it, and received the backing of the Commons to do so, not just once but twice, with a second (weaker) version of the duty proposed by the House of Lords again rejected when the Bill returned to the Commons.¹³² A last-minute concession offered by the Government will see a consultation take place on ‘enhancing information sharing between relevant agencies or public bodies to help identify and mitigate the risks of foreign interference in political donations that are regulated by electoral law’,¹³³ but there is no commitment to do anything more than to table a report setting out conclusions and next steps.¹³⁴ The possible route of foreign money into the democratic process therefore remains open, notwithstanding the numerous arguments – based on considerations of both democracy and national security – made for closing it off, and that the ‘pedigree’ of the persons making those arguments was, it was observed in the Commons, ‘phenomenally strong’.¹³⁵ The Act, that is, closes off the possibility of more blatant forms of interference but leaves the British democratic order open to some which are more subtle and so more insidious.

Another possible source for foreign interference in the democratic process to which attention has been paid in recent times is All-Party Parliamentary Groups (APPGs) – ‘informal cross-party groups’ organised around particular issues or topics.¹³⁶ There are several hundred of these groupings, with the number having risen significantly in the recent past, and many of them relating to particular countries. These too provide a possible route for foreign money into the political system: though there are rules governing such groups, which require (amongst other things) that they be transparent about their funding, there are currently no restrictions on where such funding may originate.¹³⁷ In recent times, APPGs have begun to attract a greater level of attention than had previously been the case: so, for example, a 2022 report by the Committee on Standards concluded that ‘the risk of improper access and influence by hostile foreign actors through APPGs is real, though difficult to measure’.¹³⁸ It is clear, however, that the risk is not merely theoretical: in 2022, MI5 issued a warning about ‘political influence activities’ being undertaken in the UK by a Chinese lawyer on behalf of the Chinese Communist Party. One vehicle for such influence activities was the ‘Chinese in Britain’ APPG, which she was ‘instrumental’ in setting up.¹³⁹

Amongst the possible measures the Committee on Standards floated (without recommending) in response to this issue were that APPGs ‘should be required to register whether a foreign government or organisation closely associated with it is the eventual funder of any benefit or benefit in kind (and if so which government or organisation), and conduct due diligence to this effect’¹⁴⁰ and bans on APPG secretariats funded or provided by foreign governments and on MPs accepting overseas trips (whether or not via APPGs) paid for by foreign governments.¹⁴¹ An amendment to the National Security Bill when it was before the Lords deals, in part, with this danger, making clear that the offence of general foreign interference can be committed by actions which may have the effect of interfering with whether or how a person participates in (amongst other things) the activities of an ‘informal group consisting of or including members’ of one of the UK's various legislatures.¹⁴² The underlying intention, it is clear, was to capture not only APPGs but also the various ‘friends of’ groups which have proliferated in recent years.¹⁴³ Consideration of the regulation of APPGs also continues outside the specific context of the 2023 Act, with the Commons and Lords Speakers recently arguing that the barrier to the creation of APPGs is too low, and that there should be a designated gatekeeper for the creation of new groups, who should consider whether their creation was justified and how they would be funded.¹⁴⁴

Other possible sources of foreign interference in the democratic process relate not to the formal elements of that process but rather the wider, informal, landscape within which the formal political processes – elections, votes in Parliament and so on – take place. For that reason the threat (if any) they pose in terms of hostile interference in the democratic process is more amorphous, and the form and legitimacy of government intervention in turn rather more complex and contestable. One of these is the numerous think tanks which now form a part of the political landscape in the UK, some of them with charitable status. Some of these are exceptionally influential, being linked either with political parties or particular factions thereof. Amongst this group those are a number which do not fully declare the sources of their funding,¹⁴⁵ some of which are known to take funding from abroad. In most discussion of this point the focus has been on the manner in which think tanks might therefore operate so as to (try to) sway political opinion and decision-making in favour of those corporate interests which fund them. If that principle is accepted, however, it seems clear that the apprehended danger must also encompass hostile states, who could in this way interfere – indirectly, but not ineffectively – in the democratic process.¹⁴⁶ The subject of think tanks was raised in the context of the National Security Bill, with Lord Wallace of Tankerness querying why, if governing political parties were to be covered by the Bill, think tanks – which, he claimed, often act as ‘intermediaries’ for states – were not.¹⁴⁷ Though the effect of the change made to the FIRS following its introduction is that foreign think tanks (and other foreign organisations) will not have to register their lobbying activities in the UK (as had been the case under the scheme as first formulated), the response to this query noted that political influence activity of think tanks would indeed be caught by various elements of the Bill, including the primary tier of FIRS, if carried out at the direction of a foreign power.¹⁴⁸ Simply funding a think tank or similar body, with no attempt to direct its activities, remains permitted.

Conclusion

In rebalancing modern national security law away from the question of terrorism and back to that of ‘state threats’, the 2023 Act returns us to a state of affairs that existed prior to the end of the Cold War, though with a number of important distinctions. One is that the renewed legal attention to ‘state threats’ encompasses a much broader ranger of threats, which go far beyond traditional espionage. Another is that it now benefits, if that is the correct word, from the experience of more than two decades of legal responses to terrorism, in the form of a variety of civil mechanisms, backed by criminal sanctions, aimed at hostile state actors. A third relates to the introduction of a foreign influence registration scheme that has no analogue in the history of national security law in the UK. For many of the most important forms of foreign interference it is no longer the case, as it was when the ISC produced its Russia report, that they are lawful in the UK, and even much of that which remains lawful is only contingently so, where rules around registration and declaration are complied with.

Nevertheless, the process of the NSA 2023's enactment suggests some difficulties with the project of which it is the centrepiece. It reflects, that is, an attempt to draw a difficult and perhaps untenable distinction between protecting the democratic process against the threats posed by hostile states and protecting its more general integrity. It seems likely that in practice that distinction collapses: the democratic process can be successfully protected against such hostile states if its more fundamental integrity is first ensured, and in the UK this is not obviously the case. Certainly, a number of possible routes of foreign money into the formal democratic process and the wider political environment remain open – some of them as a result of conscious and unconvincing decisions made in relation to the content of the 2023 Act. Some reasons for this might be suggested. One is that the SIAs themselves are hesitant to acknowledge any role for them in protecting the democratic process,¹⁴⁹ but this can only be a partial answer. The proposals made by the Lords – and rejected by the Commons – sought to exclude money originating from foreign powers from the democratic process without involving the SIAs in any capacity. Another, perhaps, is that some within the political system – whether the narrow, formal process or the wider political landscape – benefit from a situation in which money from dubious sources can enter it, and do not wish that situation to be brought to an end. Whatever is the case, it seems clear that the National Security Act 2023 responds only to some of the problems which exist in the contemporary UK democratic order, and while the rest of them are left unaddressed it may be that the problem is made worse rather than better. That is, the sort of threats to the democratic process, including from hostile states, which are not addressed here are more fundamental and therefore more insidious than are those addressed by the Act – capable of distorting the political system as a whole rather than particular moments of decision-making. We can be confident, therefore, that we have not heard the last of these issues.