The Policy Exchange Paper on the Attorney General’s New Legal Risk Guidelines: Excited Adjectives, Unpersuasive Analysis

One of the best lessons I was given by my pupil supervisors in drafting written submissions was to go back and review the use of every adjective. As they pointed out, there is always a temptation in written advocacy to use excited adjectives, but the cumulative effect often makes the submission less, rather than more, persuasive.

That drafting advice is one that could usefully have been heeded by the authors of the recent paper by Policy Exchange (“PE”) attacking the new Legal Risk Guidelines for government lawyers published by the Attorney General, Lord Hermer KC (“the 2024 Guidelines”). The PE paper is replete with excited adjectives: “extraordinary”; “likely to cause serious confusion”; “dramatic”; “constitutionally dubious” and “highly novel” (which, in the best Whitehall tradition, is not meant as a compliment).

However, the paper fails to justify the use of those excited adjectives.

The paper starts by explaining the purpose of the guidelines, citing a recent paper on the 2024 Guidelines by Sir Jonathan Jones KC, who was the Treasury Solicitor when the predecessor but one of the 2024 Guidelines was published in 2015. The purpose of the guidelines is to structure the legal advice-giving work of government lawyers, help address the perception that government lawyers are risk averse and the tendency to blame legal advice as a reason for not proceeding with policies, and to bring some consistency in the way that legal risk is described and conveyed to clients across different government departments (an aspect of the guidelines that, as a long-serving member of the Attorney General’s panels at the time, I found particularly helpful).

The PE paper then discusses the 2022 version of the Guidelines issued by Suella Braverman as Attorney General. At this stage there is a notable absence of excited adjectives: for example, the authors note without any adverse comment the passages in the Braverman guidance that encouraged lawyers, even in cases of high legal risk, to put the welcome news first (“there is an arguable legal basis for this”) and to relegate the unwelcome news (“you are very likely to lose in court if you do this”) to further down the submission – a suggestion that carried an obvious risk that Ministers would seize on the welcome news and fail to pay attention to the rather important unwelcome news. The PE paper also passes silently over the suggestion in the Braverman guidelines that the absence of any likely legal challenge was a matter that could properly be taken into account in deciding whether a policy was lawful. (As anyone who has ever advised government knows, there a very many cases where, for various reasons, it is unlikely that even a fairly obviously unlawful government decision will be challenged in court, including that those affected do not want to damage their relationship with government.) It is a welcome feature of the 2024 Guidelines that it expressly requires (§6) that the unlikelihood of challenge should be ignored in assessing the risk of unlawfulness (though the Guidelines later state at §11 that the likelihood of challenge can be considered as part of the overall assessment of whether to proceed).

The paper then launches its attack on the 2024 Guidelines.

The paper accepts that “at first sight” there is no change in the level of legal risk at which Ministers should be advised that the proposal is unlawful (“no respectable argument” in the 2022 Guidelines being replaced by “no tenable argument that could be put before a court” in the 2024 Guidelines).

However, the paper then criticises what is described as a “highly novel” development in §20 of the 2024 Guidelines, which deals with the next level down in terms of legal risk, i.e. the situation in which the best that can be said is that there is a tenable argument in favour of the lawfulness of the proposal that could be put before a court. (Another way of putting that situation is that the advice is that the government is almost certainly going to lose any challenge but that government Counsel will not be breaching their professional obligation not to advance unarguable points by taking the case). §20 of the 2024 Guidelines states that in that situation Ministers should be advised that pursuing the proposal should be a last resort and may be inappropriate particularly where fundamental rights are engaged, though §21 goes on to state it may be appropriate to proceed nonetheless in cases where the law is unsettled and a court decision would improve clarity or where a fundamental principle is at stake.

You may well think that all of that is rather sensible: the fact that that a proposal is likely to be successfully challenged in court is usually a pretty good reason for not proceeding with it. But the PE paper claims that that aspect of the Guidelines is “highly novel”, a “dramatic shift from earlier guidance” and “constitutionally dubious”. The basis for that claim is, the authors say, that “the role of government lawyers does not properly extend to instructing Ministers that it would be inappropriate to proceed with a legally high risk policy save as a last resort, or that it is inappropriate to proceed with a legally risky policy where fundamental rights are engaged. These are political and policy decisions and, as such, are the preserve of accountable elected officials and not civil servant lawyers.”

However, the use of the verb “instructing” in that statement is, at best, tendentious. The 2024 Guidelines do not claim that either government lawyers or the Law Officers have power to “instruct” Ministers to do or not to do anything. Government lawyers advise – and that is the verb consistently and correctly used in the 2024 Guidelines. But it is entirely constitutionally appropriate for the Attorney General – a government minister, accountable to Parliament – to set out the nature of the advice that he expects to be given at particular levels of legal risk.

Indeed, PE’s suggestion that it is “constitutionally dubious” for the Attorney General to do that is not just odd, but surprising coming from the authors of the paper, one of whom (Conor Casey) wrote a chapter titled “A Defence of the Dual Political-Legal Nature of the Attorney General for England and Wales” in a book (“Sceptical Perspectives on the Changing Constitution (2023)” edited by the other (Yuan Yi Zhu). In that chapter, it is noted with approval that “in recent years, Attorneys General have been at the forefront of executive engagement with the jurisprudence of the senior judiciary”. But it is harder to think of a clearer example of the way in which the Attorney General is properly exercising that function than advising – and making sure that government lawyers ultimately under his direction advise – Ministers that it is generally not a good idea to proceed with proposals which are likely to lose in court. That is so not least because of the implications of such challenges on scarce court and judicial time and because of the impact that persistently taking arguments that are no more than tenable has on the wider relationship of trust between the government and the judiciary.

Moreover, even accepting that questions as to whether it is right to proceed with a policy where there are no better than tenable arguments for its lawfulness are ultimately political rather than legal, Casey’s chapter is (as the title suggests) a strong and generally sensible defence of the political as well as legal role of the Attorney General. That makes the authors’ complaint that the Law Officers and officials answerable to them and under their direction are getting involved in non-legal but political questions rather hard to understand, particularly since – as I have already remarked – decisions to proceed with proposals on the basis of legal arguments that if challenged in court are almost certain to lose and can only be defended by arguments that are no better than tenable have wider implications for the relationship between the executive and the courts, a matter very much in the bailiwick of the Law Officers. In any event, it is hard to see why it is outside the Attorney General’s mixed political and legal role to remind his colleagues – and to ask government lawyers to remind them - of the burden in costs and time that having to challenge such proposals places on citizens whom government is ultimately there to serve: a burden that is often hard to justify in cases where any challenge will almost certainly succeed, and especially hard to justify in cases where citizens’ fundamental rights are at stake.

The paper further complains that it is a “novelty” (again not intended as a compliment) for the 2024 Guidelines to tell government lawyers to refer to their line managers (who should consider involving the Law Officers) cases where Ministers propose to proceed on the basis of arguments for lawfulness that are no better than tenable: §22 of the 2024 Guidelines. The paper claims that that proposal will increase the extent to which the Law Officers become personally involved in giving advice. But quite why advice being given by an Attorney General accountable to Parliament rather than by a government lawyer who is not so accountable is supposed to be a constitutional problem is not explained: and the claim is particularly odd given Casey’s defence, in his chapter, of the benefits of the Law Officers being members of, and accountable to, Parliament, rather than mere technically competent lawyers. In the end the complaint seems to be that the result will be that the Attorney General will wield a “de facto veto” over a wide range of policies: but the assertion of a “veto”, or even a “de facto veto” is not explained.

The rest of the PE paper complains about the emphasis placed in the 2024 Guidelines on compliance with international law. The political background to this – the last government’s decision to propose to Parliament legislation that it admitted was a breach of international law, and the general attitude of contempt for international law that has become too prevalent in the Conservative Party and other right-wing parties – is well-known: and the Attorney General explained in his Bingham lecture his concern as to the reputational damage that that caused (because of its political content, that passage is redacted in the published version).

None of that background is recognised by the PE paper. Instead, it begins by complaining that the 2024 Guidelines underplay the point that breaches of international law often carry little litigation risk and that they specifically state that “policies or actions which have little or no chance of being tested before a court and which are assessed as carrying a high risk under international law should be scrutinised very carefully by government lawyers”. However, as the paper acknowledges, that passage is followed by a passage that explains the type of analysis that is required in such cases: a proposal that is highly likely to be in breach of international law “may incur significant consequences, be they legal, political, diplomatic and/ or reputational”:

“An assessment of the risks of a breach of international law will require legal and policy assessments of the reputational, diplomatic and Parliamentary impact to be put clearly to Ministers. It will also require an assessment of the likely response of the actors to which the UK owes the international obligation, the likely response of the international community as a whole and any broader implications for the application and development of international law.”

Quite what is said to be wrong with that reminder of the potential repercussions of the UK being seen to be in breach of its international legal obligations is not clear. The complaint appears to be that government lawyers are not best-placed to assess such matters: but since there is nothing in the Guidelines that suggests that government lawyers will not here play their usual role in cases of complexity and expert judgment of ensuring that appropriate expertise is engaged and taken properly into account rather than trying to act as experts themselves, that complaint is without foundation. Nor is there any merit in the complaint that the 2024 Guidelines fail to recognise the different nature of international law as compared with domestic law: the audience at which the Guidelines are aimed is, after all, fully aware of those differences and does not need them to be spelt out.

The excited adjectives return in force when the Policy Exchange paper considers the last part of §9 of the 2024 Guidelines, which states that:

“the rule of law requires compliance by the state with its obligations in international law as in national law, even though they operate on different planes: the government and Ministers must act in good faith to comply with the law and in a way that seeks to align the UK’s domestic law and international obligations, and fulfil the international obligations binding on the UK. To honour the UK’s international obligations, the government should not invite Parliament to legislate contrary to those international obligations.”

That passage is described by the PE paper as “a new and radical constitutional principle” (again, not intended to be a compliment). After a brief detour in which we are reminded of the dualist approach of the United Kingdom to international law, it makes two criticisms: both of these are embellished with excited adjectives, but each boils down to the complaint that the final sentence is a “novel and extraordinary” limitation on Parliament’s power to legislate in breach of international law and on Ministers’ powers to propose such legislation.

But that criticism is bizarre. It is obvious – and scarcely needs spelling out – that a document such as this is incapable of imposing any legal limit on what Parliament can legislate or even on what Government can propose as legislation: it is and can be no more than a policy statement of intent (a policy statement entirely in keeping with the Attorney General’s political as well as legal function).

The paper then goes on to complain that the 2024 Guidelines do not expressly deal with the situation where domestic and international law are in conflict. But, again, that possibility is a matter about which the addressees of the 2024 Guidelines will be well aware and is to some extent addressed in the passage quoted (the verb “seeks” in “seeks to align” being an implicit acknowledgement that alignment may sometimes not be possible). It may fairly be said that the 2024 Guidelines take the same approach to that difficulty as the Minister’s private secretary did in “Yes, Minister” when asked who, between his Minister and his Permanent Secretary, he would obey when the chips were down: he responded that his job was to ensure that the chips stay up. But the claims in the PE paper that that “Yes, Minister” position involves some “extraordinary” constitutional heresy is, to put it mildly, overwrought. Even more overwrought is the final suggestion in the paper that this alleged constitutional heresy opens the way for Ministers to by-pass Parliament by entering into treaties that override domestic law: a suggestion that is without any foundation in the 2024 Guidelines, which do not assert (and could not assert) any such power. Rather than tilting at that windmill, the authors’ concerns about the executive using treaty-making powers to by-pass Parliament would have been better focused on considering legislation such as section 2 of the Trade Act 2021 and section 31 of the EU (Future Relationship) Act 2020, both passed under the last government, and both of which confer very wide powers on Ministers to amend domestic legislation in the light of treaties the executive has entered into and to do so with minimal Parliamentary oversight.

I noted at the beginning of this blog that the PE paper quotes Sir Jonathan Jones’ paper on the 2024 Guidelines. However, despite Sir Jonathan’s enormous experience of advising government on difficult policy and litigation issues and his practical experience of operating the previous guidance, the PE paper then ignores the thrust of his paper: which is, in my view correctly, to argue that the 2024 Guidelines are a relatively minor reversal of unfortunate aspects of Braverman’s 2022 guidelines, and a welcome indication that the Attorney General is behind government lawyers when they give Ministers that legal advice which they think best. Anecdotally, the new Guidelines are also seeing a greater willingness by government lawyers in litigation before public law courts to drop bad arguments and bad cases: which in a time when public money and court resources are stretched is no bad thing. All in all, the 2024 Guidelines are a welcome reversal of aspects of Suella Braverman’s unfortunate term in office as Attorney General (a record I considered here) and a constitutionally sound statement of, and implementation of, the Labour Government’s commitment to the rule of law.