Devolution Post-EU

I suppose that there are two priority areas. First, as other witnesses to the committee have identified, strengthening the Sewel convention is fundamental, because, unless there is some protection for the devolved institutions against the unilateral exercise of Westminster sovereignty, there are no guarantees of anything. There is a limit to how strong the constitutional guarantees that one can provide under the current constitutional settlement can be, but we need to try to get back to the situation that we were in pre-Brexit, in which parliamentary sovereignty still existed in principle, but its operation in practice was constrained. That is fundamental.

The second priority area simply relates to the increased complexity of the devolution frameworks, which increasingly concerns me. They were always complex, so we must not exaggerate what has happened. There were always different types of constraints on competence that operated in different ways. They were always difficult to understand for the uninitiated—and even for the initiated. As you will all be aware, the competence limits in schedule 5 to the Scotland Act 1998, for instance, have always been complex, and understanding exactly what they mean is not easy.

Since Brexit, however, the picture has become a lot more complicated, particularly because of the new United Kingdom internal market framework and, to some extent, because of the new external trade framework, although it is much more difficult to do anything about that. The way in which the new internal market framework interacts with the devolution statutes is highly problematic, so it would be worth trying to address and simplify that.

I have seen it suggested that one way to strengthen the Sewel convention would be just to start respecting it again. Of course, on one level, that is true, but that would not provide any guarantees against a further swing back in the future, so it would probably not provide anyone with enough reassurance that things would change.

One of the big problems with the fact that the UK Parliament has acted without consent for the first time and has exercised its section 33 reference powers and its section 35 veto power for the first time is that it then becomes much easier to do those things for the second time. Once a precedent has been established, it becomes much more difficult to stop resort to those powers in the future. To indicate a reset, it is not enough just to say, “We won’t use them—promise”. We need something more formal to provide a reassurance that things will change.

I will take your questions in order. Have the problems always existed? Have they always been latent in the system? The answer is yes. We know that devolution was enacted in the context of the continuing sovereignty of the Westminster Parliament. Various powers were written into the settlement, which is highly asymmetric: the Scottish Parliament is subject to limits that the UK institutions are not subject to.

There was real ambiguity in the initial settlement. On the one hand, there was the legal position, and, on the other, there was the political narrative: the Sewel convention is very important and, although sovereignty exists, it will not be exercised normally—whatever “normally” means—except with consent. The UK Government said, “We have these powers, but they are nuclear options. We don’t want to have to use them if we can absolutely avoid it”. There was a dual narrative.

Devolution also took place in the context of a range of other constitutional reforms that were introduced by the new Labour Government and that put parliamentary sovereignty under pressure. From about 2000 onwards, the idea of sovereignty really came under pressure. Would sovereignty continue to exist in its current form? Eventually, would it have to give way to the new constitutional reality, in various ways? There was the idea of constitutional statutes. The Scotland Act 1998 is a constitutional statute; it cannot be impliedly repealed, for instance. There were various other things, so sovereignty was under pressure from not just devolution.

The situation is very ambiguous. We might call it constructive ambiguity, a deliberate abeyance in the constitution or a deliberate constitutional silence over what exactly the new constitutional development means.

With Brexit, we saw a sweeping aside of all that ambiguity and a reassertion of parliamentary sovereignty, but not just in its traditional form. One of the very interesting things in the courts is the way in which sovereignty has been extended in the past few years. We had a reversion to a traditional understanding of the constitution, with bells on. That places the devolved institutions in a very precarious position. Brexit has been an important change.

Donald Cameron said that the use of section 35 in relation to the Gender Recognition Reform (Scotland) Bill was nothing to do with Brexit. Yes and no. Of course, the bill has nothing directly to do with Brexit, but I identified in my submission the indirect effects of Brexit, one of which has been a resurgence in what some people call muscular unionism or assertive unionism. We have to see the use of the section 35 order very much in that context. It is impossible to believe that, in the preceding two decades, no legislation was passed by Holyrood or any of the other devolved Parliaments that did not have a knock-on effect on reserved law, but we know that those knock-on effects were dealt with in different ways. They were dealt with through section 104 orders, for instance. In some cases, they were dealt with through UK primary legislation.

That gets us to Donald Cameron’s final question. Is section 35 an inevitable safety net? It is worth bearing in mind that the position is not the same under the three devolution statutes. The provisions operate completely differently in relation to the Northern Ireland Assembly. In that case, it is not the Presiding Officer who refers a bill for royal assent but the Secretary of State for Northern Ireland, and there is a different set of restrictions on when the secretary of state can recommend the withholding of consent. The position is also different under the Government of Wales Act 2006.

Is the provision necessary? The fact that it is not the same across the UK might tell us something important, but, almost always, there are other things that can be done, and should be done, to avoid the use of the section 35 veto power being necessary. I worry about that power being used simply for policy reasons—the Gender Recognition Reform (Scotland) Bill falls into that category—because there will often be these kinds of knock-on consequences. The choice not to deal with those knock-on consequences in some other way but to veto has to be seen in the light of the policy difference in relation to gender self-identification.

It is worth saying that the Gender Recognition Reform (Scotland) Bill was a great context for the first use of section 35, because it is such a controversial bill. If gender recognition reform had not been controversial in Scotland, I cannot imagine that section 35 would have been used. There was an opportunity for the first use of the power, and that opportunity was taken. It was taken late in the day. I do not think that it was on anyone’s radar until relatively late in the day, but it then became the preferred option. For most of the time, opponents were talking about judicial review and challenging the bill in the courts. Use of section 35 came to light as a possibility relatively late in the process.

It is not available for use in all circumstances. You have probably seen as much as I have people asking for section 35 to be invoked right, left and centre. It is not available generally against devolved legislation; it is available only in a limited range of circumstances. The particular part that was used in relation to the Gender Recognition Reform (Scotland) Bill is for devolved legislation that modifies the law as it operates in relation to reserved matters and has

“an adverse effect on the operation of the law”.

09:45

There are two areas of ambiguity. It is not entirely clear what kind of laws are caught by that—in my view, it would be worth clarifying that. The other issue is what standard of review the courts would adopt if the use of a section 35 order were challenged, either in this case or in the other two situations, which are an impact on international obligations or an impact on national security.

It is difficult to know, because different kinds of consideration could be in play. On the one hand, a court might say, “This is high politics. It is about relations between political institutions. We should be wary of intervening here.” On the other hand, at the other extreme, the court might say, “This is an executive actor using a power to veto, or at least require reconsideration of, primary legislation passed by a democratically elected legislature, and therefore we should scrutinise that decision carefully.” You can see both sets of arguments being made, and both sets of considerations are valid. Where a court would land on the standard of review question is therefore difficult to predict.

As far as I am aware, the keeping pace power has not been used. When I appeared before your predecessor committee, I was quite critical of the keeping pace power when that bill was being considered, because there are problems in conferring that kind of power on ministers.

However, there are obvious problems to do with the opportunities for divergence in the current constitutional landscape. The Scottish Government and Parliament, whether acting through primary or secondary legislation, always have to be mindful of their ability to diverge effectively from UK legislation. It is not about whether they can do so legally—yes, they can—it is about whether they can make that policy or legislation practically effective. In situations where the UK Government decides to take a different line for England, it will often be difficult for the Scottish Parliament and ministers, or the Welsh Parliament and ministers, to effectively maintain a different policy, whether that is a policy to keep pace with the EU law or to diverge from it in some other way.

I therefore do not think that there is anything peculiar or special about the keeping pace power, but there is a general problem with the ability to diverge effectively in many situations. It will not be the case in every situation, because a lot will depend on the precise market in which you are trying to intervene, the state of trade, the balance among local producers, imports and so on but, very often, it will be problematic.

The approach that the UK courts have adopted on Northern Ireland is a little less clear but, at least for Scotland and Wales, the approach is to treat the devolution statutes the same as any other statute. The main focus is on the words of the statute understood in their context. The courts have largely, but not entirely consistently, eschewed the idea that they should read in additional interpretative principles or understandings of what the devolution statutes mean.

That approach is justified mainly on the basis that it is the most predictable and that it preserves the impartiality of the courts when they are dealing with conflicts between Westminster and a devolved legislature. That approach works well if the devolution statutes are amended only in a consensual manner. If the restrictions that are in the devolution statutes have been put in there following a considered process with the agreement of UK and devolved Governments and legislatures, that approach is defensible but, once the consensual development of the devolution statutes breaks down, it becomes an approach that inherently favours Westminster.

We saw that clearly illustrated in the UK Withdrawal from the European Union (Legal Continuity) (Scotland) Bill reference in 2018. If you remember, after the bill had been referred to the Supreme Court but before the Supreme Court decision was made, the Scotland Act 1998 was amended without the consent of this Parliament in a way that rendered a bill that had been largely within competence largely outwith competence. All that the court did was look at the words of the statute, so there was no pushback or defence that the court could offer against that shifting of the goalposts—and it was a shifting of the goalposts.

The other issue in recent years has been the use that the Supreme Court has made of section 28(7) of the Scotland Act 1998. As you know, that says that the power of Westminster to continue to make laws for Scotland is unaffected. We used to think that that was a purely declaratory provision that had no significance at all. However, according to the Supreme Court in the UK Withdrawal from the European Union (Legal Continuity) (Scotland) Bill reference and the references of the United Nations Convention on the Rights of the Child (Incorporation) (Scotland) Bill and the European Charter of Local Self-Government (Incorporation) (Scotland) Bill, that section acts as an additional constraint on devolved competence.

It is very difficult to understand how the court gets to that position, but it seems to be relying on what it says it should not be doing, which is the reading-in of things that are not in the Scotland Act 1998. One of the difficulties is that, say, in the UNCRC case, the court objected to this Parliament trying to impose interpretive duties on UK legislation in devolved areas. There is no doubt that this Parliament can repeal or amend UK legislation in devolved areas, so it is difficult to understand why you should not be able to say, “Well, we want the courts to read UK legislation differently.” In effect, it is an amendment process by a different route. The court, however, was not prepared to allow that and seemed to be relying on reading into the Scotland Act 1998 a non-delegation doctrine. It did not say so explicitly, but that seems to be what is going on there.

The courts are, therefore, not consistent in that approach of relying purely on the wording of the statute—they have a set of background constitutional principles in mind. You may also remember the AXA case, in which the Supreme Court held that, at common law, Holyrood was subject to certain constitutional constraints—rule of law and human rights constraints. Therefore, there is some supplementing going on, but it concerns a set of constitutional assumptions, principles or values that are not specifically tailored to devolution.

I am suggesting that it may now be necessary to require the courts to pay attention to the specific constitutional context and requirements of effective devolved lawmaking. To go back to the UNCRC case, for instance, the Supreme Court was very concerned about maintaining the effectiveness of the UK Parliament’s ability to legislate in devolved areas, free from any kind of constraints—political constraints or whatever—imposed by the devolved legislatures. However, it did not have any apparent concern for the effectiveness of the Scottish Parliament’s lawmaking.

In the context of the UNCRC case, if you limit the application of the UNCRC to devolved legislation affecting children, you are missing out large parts of the devolved statute book, so you are creating a problematic patchworky approach. That concern for good devolved lawmaking was absent from the Supreme Court’s decision.

Similarly, in the case of the UK Withdrawal from the European Union (Legal Continuity) (Scotland) Bill reference, the court objected to this Parliament trying to impose a consent requirement on UK ministers when they are making delegated legislation in devolved areas. Again, the court saw that as placing limits on the effective exercise of the UK Parliament’s legislative competence. That is not really what it was about; it was about ministerial lawmaking, which is not sovereign. It is about the ability of the devolved institutions to try to maintain some control over what happens in the devolved policy space. As you know—the committee made this point in previous reports—the ability of this Parliament to have any kind of scrutiny function in relation to UK ministerial powers is entirely dependent on the quality of consent or consultation required with the Scottish ministers. Again, there is no concern for good governance at the devolved level.

That is what motivated the recommendations that I made. The courts need to be forced to take account of good governance at the devolved level and to not see devolution purely through a traditional Westminster-centric constitutional lens.

10:00

That is the thinking, but there is no guarantee with such an approach. When you create broad interpretive principles, they still have to be interpreted by the courts; the courts still have to decide what kind of weight or relevance they will give them, and they might not interpret them in the way that you anticipate.

However, I do not think that there is any hope of reorientating the approach taken by the Supreme Court without legislative intervention—at least, not with things the way that they are at the moment. Since Brexit, the Supreme Court has reverted very much to a traditional understanding of the constitution, so I think that it will only respond to clear statutory directions from Westminster.

Do you mean in preventing the Scottish Government from taking too many powers?

One criticism that has been made of Holyrood—I do not know to what extent it applies to the other Parliaments—is that the policing of the devolution settlement is very much a bureaucratic process that is done by Government lawyers, in consultation with parliamentary lawyers and UK Government lawyers, at the end of which you get a statement from the Presiding Officer that says, “This bill is compatible,” or, as has happened in the case of only one Government bill, “It is not compatible.”

There could be a role for the Scottish Parliament to play in interrogating such competence statements more fully, but the competence statements themselves would have to become fuller and better reasoned. There is a precedent at the UK level—there could be an analogy with the Joint Committee on Human Rights, which, under section 19 of the Human Rights Act 1998, scrutinises statements that Westminster bills are or are not compatible with the European convention on human rights. That committee takes evidence and publishes reports on whether it agrees with the Government’s statements. Therefore, there certainly could be a bigger role for the Parliament. The House of Lords Constitution Committee also plays that sort of role, and, of course, this committee does, too.

However, with all these things, there will always be a capacity issue. It is a common challenge for this Parliament that it is stretched. I noted that, at the committee’s previous session, Professor Nicola McEwen suggested that the Parliament should be larger, and I agree. There is a case for increasing the number of MSPs, particularly since the Parliament’s competences have expanded. Therefore, it necessarily requires more resources to ensure that those competences are properly exercised.

If you increase the opportunity for departure from the market access principles, you are necessarily “watering down”—if you want to use that language—those principles.

The curiosity about the internal market act is that it is much, much stricter than the EU rules that it replaced. I have never seen or heard a justification for why that is the case. Why, suddenly, in the post-Brexit environment, does trade need to be protected so much more strongly against other non-market objectives? There is always a balance to be struck.

The act embodies a policy choice for unrestricted free trade over other types of regulatory objectives—except, of course, it does not necessarily do so, because you have an exemption process. It is possible for the market access principles to be disapplied on an ad hoc basis, but that is unpredictable. The process is very strongly controlled by the UK Government, with limited input from the devolved Administrations. It is not that the internal market act prevents divergence or prevents a balance being struck between free trade and other objectives; rather, it just passes that to the discretion of UK ministers.

I have no insight into the decision-making process around the deposit return scheme. The process for formally requesting an exemption is, of course, not a statutory process; it sits within the common frameworks process. I do not know at what point that procedure was formalised or how widely known it is, and there is always a problem with accessibility, transparency and intelligibility when things are not contained in statutes. Maybe those were considerations. There might still be some uncertainty about whether an exemption would be required. Internal market act rules do not operate in the same way as the EU rules operated. We are in a new, complicated, uncertain situation. That level of uncertainty might have been a factor as well, but I do not know.

There is an interesting point about the exemption process. You mentioned single-use plastics. Yes, the process took a long time and that is one objection to it, but it is also really restrictive. Instead of exempting single-use plastic products as a category, it exempts a specified list of products. As you know, people are constantly coming up with new, inventive uses for single-use plastics, so you are constantly chasing your tail on this and constantly having to go through the same process even though the principles at stake might well be the same.

On deposit return, Alister Jack has been reported as saying that the threshold for granting an exemption is a very high bar. Is it? The statute does not say so. The statute is entirely permissive, so you have a UK Government minister deciding that the threshold for the grant of an exemption is a high one. A different minister might take a completely different view in a different context, depending on their view of the merits of the exemption being sought.

There is a dispute resolution procedure in the common frameworks process that, as far as I am aware, has not yet been used. There are general dispute resolution procedures under the new intergovernmental relations machinery. Again, I am not aware that those have been invoked yet. There are informal political dispute resolution mechanisms that are as yet untested, and then there are judicial mechanisms. It is always possible, of course, to seek judicial review of a decision, but it is harder to judicially review the exercise of a power that is so broad in its specification. The broader a discretion is, the harder it is to challenge that.

I had a look at the relevant common framework, which is waste and resources, I think. That is still a draft common framework. It refers to deposit return schemes as one area where divergence might be expected. However, it is very vaguely worded, and it would not be enough, I do not think, to found a claim for legitimate expectations, for instance, which might be a possibility in other contexts. It is difficult to see how you would mount an effective legal challenge to the failure to exercise a discretionary power, and, as I said, the other non-legal dispute resolution mechanisms are, as yet, untested.

Common frameworks, as I understand them, have not been used in the way that we might have expected them to be used. They were envisaged as a sort of harmonisation process. That would be the context in which you would decide when it was necessary to have a common set of rules and when divergence was acceptable. The internal market act was plonked on top of that consensual, negotiated harmonisation process, and my impression is that that sort of derailed the common frameworks, because the point of the common frameworks then became unclear. They are all a lot vaguer and more process-oriented, rather than substantive, than they might otherwise have been.

Could you make them more formal? One way in which I suggested that they could be made more formal is that, if there were an agreement to diverge in a common framework, instead of that being a ground on which an exemption might be granted, it could become mandatory. At that point, decision making in the common framework process becomes much more important, as do those issues that you raised about transparency, participation and so on. Any degree of formalisation would raise concerns about shadow or parallel law making, but that is in the nature of any kind of intergovernmental process. Those kinds of concerns about accessibility, participation, transparency and accountability are inherent to intergovernmental negotiations and are inherently difficult to address.

I am not sufficiently up on the detail of individual common frameworks to say whether that is true or not. We are really only beginning to see the practical implications of this whole post-Brexit internal market framework play out in relation to particular decisions, so it might be too early to tell. I cannot really provide a fuller answer than that; I am sorry.

The general principles of EU law are disapplied by the Retained EU Law (Revocation and Reform) Bill. As you probably know, in EU law, human rights operate on two levels: they are operated by the Charter of Fundamental Rights of the European Union, which was never part of retained EU law; and they have an influence via the general principles, which can be used to interpret retained EU law. That will go. Anything that is retained of retained EU law, which will become assimilated law, will not include or be subject to those general principles. Therefore, there is a potential issue there, although, of course, our domestic human rights frameworks will continue to apply.

The more general problem with the cliff edge is that it is absurdly short. It is ridiculous to expect that everything that would be subject to the act can be reviewed and a sensible decision taken about its retention, amendment or replacement by the end of this year. That just will not happen.

If the bill is passed, the deadline can be and will have to be extended, but that is not in the control of this Parliament; it is in the control of UK ministers. They have the option to extend the deadline across the board or for particular areas of law. I cannot see there being a cliff edge at the end of this year. It would be ridiculous that, just because civil servants had not managed to get through the work, bits of the statute book would simply disappear. That is ridiculous, so I do not think that that will happen.

That is a big question. The Brown commission report provides an opportunity. As you will be aware, the Brown commission recommended quite significant reforms to devolution—including to the operation of the Sewel convention—which are under consultation at the moment. As I understand it, the Labour Party has talked about a taking back control bill—a nice bit of political theatre—by which it means taking back control largely to the English regions but also to the devolved institutions in Scotland, Wales and Northern Ireland.

That provides an opportunity for the devolved Parliaments, because part of the Labour Party’s commitment to reinvigorating Sewel is a commitment to doing so in a consensual and agreed manner. If the Labour Party really means that, it would provide an opportunity for this Parliament, the Senedd and the Northern Ireland Assembly to say, “If you want our consent, here are our conditions”. If Labour does not get into power after the next UK general election, it is much harder to see an opportunity, but there may be other avenues.

Your opportunity to do that is through the legislative consent process. From the outside, that seems to work pretty well in this Parliament, and certainly compared with how it may have operated in the past. Committees take evidence, there are plenary debates, and we get pretty full legislative consent memorandums and, sometimes, supplementary memorandums. If you compare that with what goes on in some of the other Parliaments, a committee of the Northern Ireland Assembly complained in a report a year or so ago that the Northern Ireland Executive did not always ask it for consent. Its procedures really do not work all that effectively. The Welsh Parliament has also complained sometimes about the quality of engagement from the Welsh Government.

The quality of the engagement that the Scottish Parliament is able to have depends on time, which will often depend on timescales at Westminster, over which you have no control. There may be opportunities for more interparliamentary working. That point is often raised. Where UK Parliament committees are taking evidence or looking at bills, there may be opportunities for joint working, or at least for a more direct feeding in of views, but I have not thought in detail about how those procedures might work.

When Sewel was put on a statutory footing in the Scotland Act 2016, I wondered whether that interpretive role was one way in which it might have had some judicial teeth. There were always going to be issues around direct enforcement. In the first Miller case, the Supreme Court said that it was not even going to get involved in interpreting the Sewel convention. It could still have had some sort of interpretative function, but I do not think that it has. You sometimes see the statutory recognition of Sewel and the referendum lock being referred to in cases, but then they play no role in the decision-making process. They are there, but they are not doing any constitutional work, as it were—certainly not in the adjudicative sphere.

10:30

Strengthening the statutory statements of Sewel could work. The Brown commission has proposed taking out “not normally” so that it looks like an absolute rule. Of course, it would not be an absolute rule because the enforcement would be via the House of Lords or revised second chamber veto. The second chamber would, in effect, be deciding what was normal and what was not normal and when it was justified to depart from the convention. That is not good enough. I would rather see a more explicit set of exceptions to the Sewel convention, and whether that is done on a statutory basis or through intergovernmental negotiation is a question for discussion. Some sort of explicit understanding of when override is justified would be best because “not normally” is far too vague.

In the Brexit process, we heard the view that Brexit was not normal. Okay, Brexit was not normal, but that did not mean that every piece of Brexit-related legislation raised the same issues, had the same urgency or was being done against the same background of external obligations. In some cases that was true but not in others. A blanket invocation that Brexit is not normal is not good enough. You need some sort of clarification of the principles. Obviously, there would still be room for discussion about whether they apply in a particular case, but you would have a clearer understanding of when setting aside consent might be justifiable.

One of the weaknesses in the initial set of intergovernmental arrangements and devolution guidance notes and so on back at the beginning of devolution is that there was very little parliamentary involvement. There was some in this Parliament, and there was none in the UK Parliament. You would not want to replicate that: you want buy-in from all four Parliaments, particularly the UK Parliament. If this is the UK Parliament subjecting itself to self-denying ordinances, it needs to know about them, acknowledge them and commit to them explicitly.

Yes, the Brown commission proposal ultimately depends on politics to defend the devolved autonomy. I wrote about that when the report was first published. That makes the composition of any new second chamber absolutely crucial, because if the devolved territories are represented in proportion to population share, there is very little guarantee. The assumption is that if you have more devolution in England, somehow that will change the general constitutional mindset and they will be more attentive to such matters, but it seems to me that there are no guarantees there.

Brown envisages a role for the courts in triggering the House of Lords veto. There would probably have to be a reference to the Supreme Court—it is not specified—to certify, as it were, that the bill in question is one that engages the second chamber veto power, so it gives the courts a role.

I can see the case for that, because you will be aware that, over the past few years, we have seen not only instances of lack of consent being ignored but many more disputes about whether devolved consent is required at all. Sometimes, it seems that the UK Government is taking quite a narrow view of devolved consent. Sometimes, it seems that the devolved legislatures are taking a very expansive view of when devolved consent is required. At the moment, there is no way of authoritatively resolving those disputes.

Miller 1—the first Miller case—was an attempt at least to get the courts to say whether article 50 legislation would engage the Sewel convention, but the courts just said, “No, we are not going to touch it.” At the moment, you have no way of authoritatively resolving those disputes, which means that, by default, the UK Government’s position prevails. You have a double whammy there. The UK Government can undermine Sewel from two directions. One is by confining its scope, and the other is by being willing to legislate even within scope, where it acknowledges that the convention is engaged, by being willing to dispense with the requirement of consent. Both of those are problematic.

I do not see how you can get an authoritative resolution to those scope problems without involving the courts. The House of Lords Constitution Committee said that Westminster parliamentary committees could perform that role, but they are political committees. The devolved Administrations are not directly represented before those committees. They could give evidence, but that would be it.

The Brown commission proposals require the second chamber to trigger the reference to the courts. As far as I can see, there would not be any role for the devolved Governments or devolved legislatures to trigger that reference procedure, so you would want to widen that out a bit. What you are talking about here is the reassurance that comes from an independent adjudicator. At least at that point of scope, that requires legal intervention.

I am not sure that it would be less contentious. The Brown commission envisages the role of the second chamber as being a much more constitutional one. Therefore, the culture of the house would be more attuned to the protection of constitutional matters. We already see that in the current House of Lords, which tends to be more interested in constitutional issues and issues of principle than the House of Commons is. However, that is an unelected house. If you have an elected and, actually, weaker house—Brown proposes removing the House of Lords’ routine delaying power and replacing it with a stronger veto power—it is not obvious to me that that is a recipe for a more constitutionally vigilant second chamber.

An English parliament is not a realistic prospect. In a sense, it is not really an English parliament that you have to worry about, it is the Government. That is where the problems stem from. It is Whitehall departments that fail to understand when they are acting with a UK hat, or an England and Wales hat. Once it gets into the legislature, it usually becomes clear what the territorial extent of legislation is. As we saw with EVEL—English votes for English laws—you can introduce procedures that separate England-only legislation from UK or Great Britain-wide legislation. However, it is much harder to see how you can do that at Government level, and it is particularly difficult to see how you can do it in circumstances where it really matters; in other words, when you have different English and UK-wide political majorities, because you would have to have two different Governments. It is very difficult to solve that problem.

Does more English regional devolution solve the problem? As you know, the model of English regional devolution that is being pursued is a very different thing from devolution in Scotland, Wales and Northern Ireland. It is not legislative devolution. There are no separate representative assemblies. There are directly elected mayors, but that is a different thing. We are talking about enhanced local government, as opposed to the parcelling out of central Government authority, which is what devolution does. Devolution creates institutions of central Government with accompanying legislatures in the devolved nations. English devolution, at the moment, does not attempt to do that, and I have never seen any concrete proposals for legislative devolution in England.

10:45

While they continue to be different things, I do not think that English devolution is a solution to the constitutional problems that the Scottish, Welsh and Northern Irish institutions face. That is not to say that it is not worth doing. It tends to be perceived in a different way in England, where it is about decentralisation rather than about a legitimacy problem, which is what drove devolution in Scotland in particular.

No doubt, and you can take negative lessons as well as positive lessons. Each set of constitutional arrangements is unique, and whether they are formally federal is not necessarily as important in practice as it might appear in principle. I will leave you with a very general answer. I am sure that there are lessons that you could take, but, off the top of my head, I cannot give you any more detailed an answer than that; I am sorry.