Investigatory Powers Bill (Sixth sitting)

Debate between Keir Starmer and Suella Braverman

Committee Debate: 6th sitting: House of Commons
Thursday 14th April 2016 Public Bill Committees

[Keir Starmer]

I am grateful for that intervention. If the Solicitor General can point to such a provision, I would be interested to see it. On the face of it, the clause allows designated senior officers within a public authority to obtain communications data in the interests of the economic wellbeing of the UK without that further qualification.

Subsection (7) then states that data can be obtained

“in the interests of public safety…for the purpose of protecting public health”

or,

“for the purpose of assessing or collecting any tax”.

We then come to paragraph (g), on which I want to spend some time. It states that data can be obtained

“for the purpose of preventing death”—

that would obviously be a high threshold—

“or injury or any damage to a person’s physical or mental health, or of mitigating any injury or damage to a person’s physical or mental health”.

The threshold is way, way down. There are many ways in which a person’s physical or mental health could be damaged. The Bill, if passed, will authorise access to communications data without any threshold as to the level of damage or injury.

[Suella Fernandes (Fareham) (Con)]

I appreciate the hon. and learned Gentleman’s analysis, but does he agree that obtaining communications data is one of the less intrusive powers contained in the legislation, but such data are very helpful for setting the scene and planting the seed for investigations? That has to be borne in mind when looking at the authorisation regime, because this is different from other powers.

[Keir Starmer]

Let me take that in stages. I accept that accessing communications data is in a different category and order from, say, the interception of the contents of communications. I also accept the proposition that communications data are used in many cases involving serious crime. I will go further than that: it is rarely possible to bring and to conclude cases of serious criminality without reliance on communications data. I have no in-principle objection to communications data being made available and being used. My concern is the very low level of sign-off required to access those data and the lack of any meaningful threshold in subsection (7); there simply is not one. Whether or not a meaningful threshold is achieved by the insertion of the word “serious”, as I propose in my amendment, or some other word, if we simply say that it could be necessary and proportionate to access communications data to prevent any crime or damage, we are proceeding on a basis for which it is very hard to think of any circumstances in which it would be difficult or impossible to justify obtaining communications data. It just is not a set of thresholds.

Dealing with miscarriages of justice and situations in which a person has died and so on, and

“for the purpose of exercising functions”

are listed in the subsequent paragraphs. My central point is obvious but important. I realise how necessity and proportionality apply, but on any reading of subsection (7) there is no threshold. I think there is a risk for the Government here. I appreciate the direction of travel of the Court of Appeal, but does anybody seriously think that the jurisprudence is not going to develop to a point where there is a threshold that is thought to be appropriate? It is one thing to say that we do not necessarily have to have a threshold of serious crime, but to go from that to saying that we do not have to have any threshold at all is to invite problems, if these provisions are passed.

[Suella Fernandes]

It appears that the hon. and learned Gentleman is dismissing the necessary hurdles of necessity and proportionality in satisfying the tests. They are obviously going to relate to and be thresholds, so is it not wrong to say that there is no threshold in the clause?

[Keir Starmer]

I appreciate that the necessity and proportionality test has to be applied—in any given case there will always be an argument about whether it is necessary and proportionate—but as ever with necessity and proportionality the question is: what are we assessing necessity against and how are we arguing that it is proportionate? Is it necessary to do what? We get that only from the face of the statute. In other words, necessity does not give us anything unless we have some subject matter that it bites on, which is why the subject matter that it bites on is so important. Whether it is necessary for serious crime is one question; whether it is necessary for crime is another.

There are many, many things that one could say were necessary to prevent or detect crime. I absolutely accept that in practice those two tests are applied at all times, but the question is: what are they applied to? The question that the designated senior officer has to ask him or herself is: “Am I satisfied that it is necessary to prevent crime?” That would be good enough under the clause. It is, in principle, an inadequate threshold. I also think it will invite challenge in due course, because I do not think for one moment that, in the long run, the European Court and our courts are going to be satisfied with a scheme that does not have any threshold, even though there will be and are arguments about the precise threshold. We can see what the divisional court said in the Tom Watson case, so it is not just counsel’s argument that was never accepted by anybody. In that case in the divisional court, counsel’s argument that the serious crime threshold was an important safeguard was accepted. Thankfully, the writing is therefore on the wall if the clause is not taken back and reconsidered.

I shall move on to the second “who”. The first “who” I focused on was who can issue the necessary authorisation, which is the designated senior officer. Under clause 53(2), that person can

“authorise any officer of the authority to engage in any conduct”.

It goes from a relatively low-level authorisation to somebody even further down in the authority having to get on with the job of obtaining data.

The breadth of what can be done is outlined in clause 53(5), which states:

“An authorisation…may relate to data whether or not in existence at the time…may authorise the obtaining or disclosure of data by a person who is not an authorised officer, or any other conduct by such a person, which enables or facilitates the obtaining of the communications data”—

so it goes beyond the specific authorisation to the facilitation—

“and…may, in particular, require a telecommunications operator who controls or provides a telecommunication system to obtain or disclose data relating to the use of a telecommunications service provided by another telecommunications operator in relation to that system.”

It is a very broad provision.

That enables us to see the amendments in their proper context. There are three categories of amendment. The first category is to be taken as a set and would insert some rigour and independence into the process by requiring judicial commissioners to sign off the necessary authorisations. The second set of amendments, which we will come to in due course, seeks to amend the threshold to provide a meaningful threshold for the judicial commissioner. To call clause 53 as drafted a set of safeguards is to mis-describe the words on the page.

Investigatory Powers Bill (Fourth sitting)

Debate between Keir Starmer and Suella Braverman

Committee Debate: 4th sitting: House of Commons
Tuesday 12th April 2016 Public Bill Committees

[Keir Starmer]

The point I am making is not that that judicial commissioner could be more accountable, or that there would be some forum in which the judge could go and explain. I completely accept that that is a limitation. I am meeting the argument against this proposal, which is that at the moment the Secretary of State has some political accountability which would be reduced or taken away if this amendment were accepted.

The point David Anderson makes is that it is of course a criminal offence to disclose that the warrant has been signed, so in fact the Secretary of State could not go to the Dispatch Box even in an extreme case. She would commit an offence if she went to the Dispatch Box to be held accountable for an individual decision. That is exactly why David Anderson writes as he does in paragraph 14.56 of his report. If any other members of the Committee have found an example of a Secretary of State ever actually being held accountable for an individual warrant, I personally would like to see the Hansard report of that taking place.

[Suella Fernandes]

On the question of accountability, there is a clear line of accountability to the Executive in the form of the Intelligence and Security Committee. It is a body of reviewers—elected, accountable and within the parliamentary and democratic process—who have access to this confidential information and can review the actions under this function. That is a clear line of accountability, which exists and is exercised.

[Keir Starmer]

Again—and I will be corrected if I am wrong on this—the statutory prohibition on the Secretary of State ever saying whether or not she signed a warrant applies across the board, whether in a Select Committee or in any other parliamentary proceedings. In other words, first, she could not be asked a question about an individual warrant because there would be no basis on which it could be put and, secondly, even if it were asked she could not answer it. I take the point that is being made but, wherever the accountability is placed, to hold the idea that there is individual accountability for the hugely important decisions that are made on individual warrants is to misunderstand how the regime works.

--- Later in debate ---

[Keir Starmer]

This is where we pick up the discussion about scrutiny. As the amendments to clause 17 were withdrawn, the premise here is that of a dual function, carried out first by the Secretary of State and then by the judicial commissioners. To be clear, we welcome the involvement of judicial commissioners, and the amendments focus on their role in the process. We have had the discussion about whether the judicial commissioners should be the default decision-makers—this is a different exercise.

What is clear in clause 21(1) and (2) is that what is envisaged in the Bill is a review exercise by the commissioners. That is clear from the words “must review”. Subsection (1) states that the judicial commissioner must review the person’s—in this case, the Secretary of State’s—conclusions as to necessity and proportionality, and subsection (2) states that

“the Judicial Commissioner must apply the same principles as would be applied by a court on an application for judicial review.”

It is therefore a review mechanism, and it is a review according to judicial review principles.

Two problems arise from that. The first is that it is not, therefore, truly a double lock. A double lock denotes a decision by the Secretary of State, which survives in clause 17, and a decision by a judge—a judicial commissioner—under clause 21, but this is not that sort of double lock.

The second problem, the reference to judicial review, is equally profound. Committee members will remember my question to Lord Judge:

“Do you agree with me that as the Bill is currently drafted, it is not clear what Parliament intends”—

in relation to judicial review of warrants—

“and therefore it will fall to the judges? In other words, it is broadly enough drafted to cover a longer-arm review or a closer intense review depending on what judges decide as cases evolve. It could accommodate both approaches.”

That is the problem with judicial review here.

I will quote Lord Judge’s response, because he captures the real cause for concern here:

“I think ‘judicial review’ is a very easy phrase to use. It sounds convincing, but it means different things to different people. People say, ‘Wednesbury unreasonableness’—that was a case decided by the Court of Appeal in 1948 or 1947, and it has evolved. Personally, I think that when Parliament is creating structures such as these, it should define what it means by ‘judicial review’. What test will be applied by the judicial—I call him that—commissioner, so that he knows what his function is, the Secretary of State knows what the areas of responsibility are and the public know exactly who decides what and in what circumstances? I myself do not think that judicial review is a sufficient indication of those matters.”––[ Official Report, Investigatory Powers Public Bill Committee, 24 March 2016; c. 67-68, Q220.]

That is one of the most experienced and well-respected judges in the country indicating that in those circumstances judicial review is not a sufficient indication of the test.

Amendment 62 would require the judicial commissioner to decide for him or herself on necessity and proportionality. Amendment 89 would take out the reference to judicial review. The scheme and structure of the Bill would therefore be retained. There would be a double lock. Both the Secretary of State and the judicial commissioner must be satisfied that necessity and proportionality is made out, at which point the warrant would come into effect, unless of course it is an urgent warrant. There would be clarity about the role of the judge.

In previous exchanges, it has been accepted that the judicial commissioner will see the material that is before Secretary of State and therefore can make that decision. The lock therefore becomes what we have termed an equal lock, where both parties make a decision on the substantive merits of the case. That gets rid of the potential ambiguity with which Lord Judge was concerned. It would then be absolutely clear that this is truly a double lock. It is a simple and straightforward amendment that would bring real clarity to the exercise.

[Suella Fernandes]

I am listening to the hon. and learned Gentleman with interest, and I appreciate his exploration of the meaning of this term. What is his opinion of Lord Pannick’s assessment of the insertion of judicial review? He concludes that it is sufficient, flexible but clear and strikes the right balance.

[Keir Starmer]

I know and respect Lord Pannick hugely, but there is no guarantee in the Bill that his preferred way of approaching this under judicial review principles is the one that will be carried out in practice; he has no control over the test that will be applied. Lord Judge’s concern is that some judges may consider that this is an area where they virtually take the decision, which is what they do in certain cases involving particular human rights issues, where they get very close to the decision, while other judges will be much more deferential.

With the best will in the world, Lord Pannick puts forward the view that judicial review will work, but there is no guarantee of that. Unless it is set out in the Bill, the test will be simply left to be applied on a case-by-case basis. Nobody, in this formulation, could argue that a judge who applied long-arm reasonableness was acting in any way other than in accordance with the test.

Obviously, I respect what Lord Pannick says, but Lord Judge was making a different point that goes back to accountability, to some extent. He was alive to the fact that once judges are involved in the decision-making process, a torch will be shone on them in relation to these warrants. There will be inhibitions on what they can say and the circumstances in which anybody could hold them to account. We have rehearsed that. I read into his answer that he wanted absolute clarity and a tightness of test so that the judges knew what they were to do and could operate within those confines, thus protecting themselves from the suggestion that they had applied too close or too loose a test. It is partly about clarity, with one eye on judicial accountability in the longer term for the decisions that have to be made.

--- Later in debate ---

[Suella Fernandes]

It is an important issue. Evidence to the Joint Committee from Sir Stanley Burnton and Lord Judge was unequivocal, in that Wednesbury unreasonableness would have no place in this context. That seems to be maintained by Sir Stanley Burnton in the evidence that we have received more recently. Does the hon. and learned Gentleman agree that Wednesbury unreasonableness has no role in this context, especially by virtue of reference to necessity and proportionality?

[Keir Starmer]

The reference to proportionality and necessity does not help in this context, because the question for the judge on this formula is not, “Is the measure necessary? Is it proportionate?” Judges often make, and are well used to making, that decision. The decision for them on this formula is whether, when the Secretary of State decided that it was necessary and proportionate, she was exercising her powers in a way that cannot be questioned, applying the principles of judicial review. That is the real difference.

Whether I think the long-arm Wednesbury test is appropriate is neither here nor there. So long as we have clause 22(2), it is open to a judge to apply the old-fashioned Wednesbury test, because that is within the principles of judicial review. The case law obviously varies. The closest possible scrutiny is usual in control order or TPIMs cases, but there are many other examples involving national security where the judges have persistently said that long-arm review applies. There are two strong lines of case law, and I am arguing that one is better than the other. The point is whether the Bill is clear enough about the test to be applied.

This is a real opportunity, as much as a challenge, for the Government. The provision is a new one, and it is a double lock if properly applied. It ought to be substantive. The judge ought to decide whether a warrant is necessary or proportionate. As long as he or she does, the warrant comes into existence and can be relied upon. In the 21st century, that is the right approach when such a provision is going into statute for the first time.

--- Later in debate ---

[Keir Starmer]

There is. That is a very good point, and it is one that I have discussed with the Bar Council. In those circumstances, what is being targeted is material that is not legally privileged, though there might be something that is legally privileged within it. There should be safeguards put around that, and I readily accept that examples will arise, probably also in the bulk powers, in which, although the intention is not to target legally privileged material, it is very difficult to have a warrant which does not run the risk.

An example would be when there is a suspicion that a lawyer and client may be involved in some activity that would take the communication outside of legal privilege, but it is impossible to say at what point of the conversation or exchange it loses its legal privilege. That is an obvious example. The answer that the Bar Council gives to that, and that I agree with, is that in those circumstances, rather than having a warrant to target the legally privileged material, there is a regime that recognises that it may be that, when targeting what can legitimately be targeted—namely, the part of the communication that has lost its privilege—there is a risk that privileged communications are incidentally picked up. There should be a provision for dealing with that material and its disclosure.

The powerful point about subsections (1), (2) and (3) is that it is wrong, in principle, to target legally privileged material. It is possible to have a warrant that runs the risk, with a separate set of safeguards to ensure that, if the risk materialises—as it will in some cases—there are provisions for ring-fencing, safeguarding, and not disclosing that material. That is the intention behind the Bar Council amendment.

It may be that further tweaks or improvements can be made, but that is an important point of principle that I invite the Solicitor General to take away and consider. A clause that satisfied the Bar Council in terms of the legal protection of this important privilege would be a prize worth having. Although the Bar Council recognises, as I do, the movement that the Government have made here, they simply have not got this right, for the reasons that I have outlined.

Subsections (4), (5), (6) and (7) are focused, in a sense, on communications that are likely to include items of legal privilege, such as a warrant that touches on a solicitor or lawyer communicating with clients, where it is thought that privilege has been lost but also elements where it has not been lost. In those circumstances, the Bar Council’s view and my view is that what is set out is again simply not strong enough, because there is no test or special provision.

New clause 2 is a comprehensive clause that would deal with that issue. In a sense, it goes with amendment 80, which amends a much later provision. It is intended to tidy up and clarify what the Bar Council says properly represents legal privilege and a regime for protecting it.

[Suella Fernandes]

Does the hon. and learned Gentleman not think that there is a special level of safeguard incorporated in the clause? A higher bar needs to be overcome. Only in “exceptional and compelling circumstances” will privilege be circumvented. Is that not a high standard to meet?

[Keir Starmer]

I accept that it is a high standard to meet, but it is focused on the wrong target. If it is wrong in principle to target legally privileged material on the basis that that material might involve communications that further crime, on a proper understanding, that material has already lost its legal privilege. Having a higher test to target something that has not lost its legal privilege is a good thing, but it is not enough. Material that has not lost its legal privilege should not be targeted, because it is in fact not furthering crime. The proper way to deal with it is to recognise that what one really wants to target is communications that have lost their privilege. However, there is a risk of including—unintentionally, because one does not want to target it—other material, and that requires a different approach and a different regime. That is really the point. It is good to have a threshold, but the threshold does not work within the confines of this scheme.

I urge the Solicitor General to view the clause in that light and to reflect again on it. A lot of work has been done to try to get it into a better state, but that has not met with the approval of the Bar Council and, following analysis and discussion with the council, I can see why. New clause 2 is the council’s attempt to get it right. It has spent a lot of time on it and is very concerned about it. I invite the Minister to reflect again and commit to looking again at the clause, perhaps with us and the Bar Council, to try to get a clause that meets with the approval of everyone concerned. If that can be achieved, it will be a prize worth having; if it cannot, it will be a waste of a bit of time on a good cause.

draft Regulation of Investigatory Powers (Interception of Communications: code of practice) order 2015 Draft Equipment Interference (Code of Practice) Order 2015

Debate between Keir Starmer and Suella Braverman

Thursday 7th January 2016 General Committees

[Keir Starmer]

I am grateful for that.

Regarding the remaining tricky or more complicated areas, I shall focus on legal professional privilege and the protection of communications involving confidential journalistic material and other confidential information. Before I do so, though, I highlight the point made by a number of respondents to the consultation on the equipment interference code. The Government’s response to the consultation summarises their point as saying that

“a code of practice was not a suitable vehicle for setting out the power to conduct equipment interference and that it should be provided for in primary legislation. This would offer an opportunity to have an open and transparent debate about the use of equipment interference by the Security and Intelligence agencies.”

That is a point well made in the consultation, although the Government’s response is inevitably constrained by the legislation that is currently in place. Nevertheless, it emphasises the need for a real debate on this issue as the draft Investigatory Powers Bill goes through its various stages.

I will not take up time by reminding the Committee of the importance of legal professional privilege, but the need for reform and further guidance under the code is absolutely clear. In that respect, probably the only quarrel I have with the Minister is that I am not sure that the new codes are simply about the Government doing their job properly. They were necessary as a result of the ruling in the Investigatory Powers Tribunal, which declared in February last year that the previous approach was not in accordance with article 8 of the European convention on human rights. That position was rightly conceded by the Government, because in that case the IPT ruled that

“the regime for the interception/obtaining, analysis, use, disclosure and destruction of legally privileged material has contravened Article 8 ECHR and was accordingly unlawful.”

It was therefore necessary, for the period that the current regime remains intact, to have further guidance to bring the approach into accordance with the IPT.

I remind the Committee that the previous code simply said that caseworkers

“should be alert to any intercept material which may be subject to legal privilege.”

It did not go on to state what steps should be taken if legally privileged material was identified. There was a deficiency there that the new code is intended to deal with.

Although they do not ring-fence legally privileged material, the new codes do provide much more detailed guidance, which, again, is welcome, particularly in paragraphs 4.5 to 4.25 of the interception of communications draft code and chapter 3 of the equipment interference draft code. I highlight the fact that the latter provides that, prior to any warrant being granted where interception of privileged information is likely, there must be an assessment of how likely it is that such information will be intercepted. So, first, there must be an assessment before the event. Secondly, when the interception of legally privileged information is intended, the threshold, as the Minister said, is that there must be

“exceptional and compelling circumstances that make the authorisation necessary.”

Thirdly, the code makes it clear that the threshold will be met when there is an

“imminent threat of death or serious injury or serious threat to national security”

but it is anticipated that such situations will be rare. In addition, the code states that any communication between lawyer and client or any third party for the purpose of actual or contemplated litigation

“must be presumed to be privileged unless the contrary is established”.

Those are three or four aspects in which the guidance is much sharper and clearer. Time will tell—in the limited life of such codes—whether the regime is robust enough. Over the coming weeks and months, we will obviously keep a beady eye on how matters progress. To some extent, however, such matters will be considered in greater detail as the Bill proceeds.

My only point at this stage is that there is a question mark over whether the protection in relation to dissemination is strong enough under the code. The code simply states that privileged information cannot be disseminated unless a legal adviser has been consulted on the lawfulness of such action and that “all reasonable steps” must be taken to ensure that “as far as practicable” authorities involved in legal proceedings are prevented from seeing privileged information relating to those proceedings. Why does the code not expressly prevent dissemination where legal advice has been received as to its unlawfulness? I accept, however, that that question is probably equally well suited to the forthcoming debate on the Bill.

Moving on, it is noticeable that the protection for journalistic material and other confidential information is a lot weaker than the protection for legally privileged material. In his report, “A Question of Trust”, David Anderson, the Government’s reviewer, points out:

“The Draft Interception Code sets out similar provisions in respect of journalistic or other confidential material but the threshold for access is not as high as that in respect of legal privilege.”

It is obviously a matter of some concern that there are two different regimes for protected information. This matter was raised in the consultation, and I remind the Committee that the News Media Association took the view that the current regulatory framework

“poses a threat to journalism, journalists and their sources”.

The new provisions in the code of course have a chequered history. The National Union of Journalists, in a joint statement with the Bar Council, said that

“access to professional data should be protected in law and should be subject to independent, judicial oversight. Using codes of practice—such as the draft code under RIPA—undermines the rule of law.”

To some extent, their plea is for a change in the law, which is hopefully now forthcoming. The general secretary of the NUJ said:

“The proposals contained in the existing RIPA code of practice simply do not offer the protection to journalists and to sources, and are in fact dangerously inadequate. New legislation is urgently needed—it is vital that judicial oversight is introduced to force police officers and other snoopers to apply to judges in a transparent process before surveillance powers against media and legal professionals can be considered.”

Finally, the Press Gazette and the Society of Editors said that the draft code provides

“wholly inadequate protection for journalists’ sources”

and demanded that communication between journalists and public officials be treated the same as privileged information.

I recognise that the target of some of those comments was new legislation rather than a different code and that the code can only go so far, but not to have aligned in the interim the protection for journalistic material and other confidential material with the protection now given in the code to legally privileged material is a missed opportunity.

[Suella Fernandes (Fareham) (Con)]

How would the hon. and learned Gentleman define “journalist” in this context given the plethora of people out there, from the occasional blogger to the editor of a mainstream broadsheet newspaper, who would self-describe as journalists?

[Keir Starmer]

That is a good question, and a difficult one to answer. I confronted it when I was Director of Public Prosecutions, because I had to issue guidance on how we would approach the prosecution of journalists. We took a broad view, on the basis that if the protection of journalists’ sources is to have any meaning, one cannot distinguish between different forms of journalism. It is simply not good enough to say that because the definition is difficult, the protection should not be afforded to any.

I acknowledge that it is difficult to define journalism. I gave it my best shot in the guidance that I published and took a broad approach, but I resist the notion that because it is difficult to delineate clearly the limits of what a journalist is, the long-standing and hard-won protection for journalists’ sources and other confidential information must yield to that difficulty. That is a dangerous path for us to go down. It is obvious and inevitable that the regime in this legislation will not involve the sort of judicial oversight that comes with the Police and Criminal Evidence Act 1984, which is a live issue in the public domain among journalists and others. As I said, I think that it is a missed opportunity, albeit for a relatively short period, not to have aligned the protections in the different sorts of protected category in the codes to give better protection to journalists, their sources and the confidential material with which they deal routinely.

There is, of course, much to focus on in the upcoming debate on the draft Investigatory Powers Bill. We welcome the codes and the tone and manner in which they have been put before the Committee. I have outlined the concerns, but we support the codes.