Written evidence from Dr Robert Brett Taylor[1] (PAC 14)

Public Administration and Constitutional Affairs Committee

Parliament and the Constitution inquiry

1. This submission intends to bring to the Committee's attention issues relating to ‘the nature and effectiveness of constitutional conventions in the parliamentary system’.
2. The following draws upon and develops points raised in: Robert Brett Taylor ‘Foundational and Regulatory Conventions: Exploring the Constitutional Significance of Britain’s Dependency upon Conventions’ (2015) Public Law 614, and ‘The Contested Constitution: An Analysis of the Competing Models of British Constitutionalism’ (2018) Public Law 500.

The nature and effectiveness of constitutional conventions in the parliamentary system

The nature of the British constitution

3. In order to understand the nature and effectiveness of constitutional conventions in the parliamentary system, one must first understand the nature of the British constitution.
4. The British constitution remains both uncodified (not written down in one place) and unentrenched (not higher law which is enforceable by the courts against all other law and not protected from change via special amendment procedures).  This is as a consequence of its evolutionary nature and the existence of the doctrine of parliamentary sovereignty.
5. As a result, the exact nature of the British constitution remains contestable, with constitutional commentators offering competing understandings of how the British constitution works or should work.  The three competing models can be briefly summarised as follows:

a)      ‘Legal Constitutionalism’ advocates the adoption of a codified and entrenched constitution which imposes legal restrictions on the powers of both Parliament and Government which should be capable of enforcement by the courts.

b)     ‘Common Law Constitutionalism’ views the English common law as interpreted by judges as already forming Britain’s de facto constitution, capable of placing limits, not only on the Government, but on Parliament also.

c)      ‘Political Constitutionalism’ propounds political accountability over legal accountability, and is thus opposed to both judicial expansionism and the imposition of any legal limits on the power of Parliament.

6. It is submitted that the British constitution is in fact a complementary fusion of all three models: what I have called ‘complementary constitutionalism’.  Furthermore, in comparison to other constitutions, the British constitution is primary political in nature. This is in no small part due to its continued dependency upon constitutional conventions, which is a fundamental feature of a political constitution but one which is also essential in striking a complementary balance.

Constitutional conventions: definition and identification

7. Constitutional conventions remain a major feature of the British constitution that are used to govern principally the behaviour of constitutional actors including the Queen, the Government, and Parliament.  Constitutional conventions are legally unenforceable but politically binding rules.  They therefore perform the same function as laws, but nevertheless remain distinct due to their legal unenforceability.
8. Historically, constitutional conventions emerged out of political practice.  As a result, they are traditionally unwritten.  Because of this, it can be difficult to identify a valid convention.  A tripartite test for identifying conventions was accordingly developed by Sir W Ivor Jennings.  Jennings’ test is as follows:

‘[F]irst, what are the precedents; secondly, did the actors in the precedents believe that they were bound by a rule; and thirdly, is there a reason for the rule?’[2]

9. Jennings’ test has since been recognised judicially both at home (Evans v Information Commissioner [2012] UKUT 313 (AAC)) and abroad (Reference re Amendment of the Constitution of Canada (1982) 125 DLR (3d) 1, Supreme Court of Canada).
10. Despite this, conventions can also be expressly created.  This happened with the Sewel convention, which was created following the introduction of devolution and was first contained in a memorandum of understanding in 2001.  An additional practice has also emerged in recent years of codifying or reducing to writing existing conventions (such as in The Cabinet Manual) as well as placing some conventions on a statutory footing (such as the Sewel convention in s.28(8) of the Scotland Act 1998 (as amended by s.2 of the Scotland Act 2016)).
11. It is submitted that reducing constitutional conventions to writing does not constitute a shift towards greater legal constitutionalism.  Such conventions, where contained in non-statutory publications or non-justiciable statutes, remain legally unenforceable. Conventions which are approved democratically may in fact have a greater chance of being complied with consistently.

The types of constitutional convention and the significance of British dependency upon them

12. It is submitted that there are two types of convention under the constitution, what I have called foundational and regulatory conventions:

a)      ‘Foundational conventions’ are those conventions which bind unelected constitutional actors, and are essential in making the constitutional system both democratic and accountable.  Examples include: the rule that the Queen exercise her prerogative powers on the advice of Ministers, the rule that the Queen will appoint as Prime Minister someone who can command a majority in the House of Commons, the rule that the Queen will give Royal Assent to Bills passed by both Houses (on the advice of her Ministers), and the Salisbury Convention in the House of Lords.  Under many other constitutions, such foundational rules would be governed by law, not convention.  Because foundational conventions in the UK bind unelected constitutional actors, they can be easily replaced with statute without undermining the British Constitution’s dependency on political accountability.

b)     ‘Regulatory conventions’ in contrast bind the behaviour of elected constitutional actors such as MPs and the Government.  They set standards by which they can be judged, thus facilitating their political accountability.  Examples include: individual ministerial responsibility, collective cabinet responsibility, the Sewel convention, and potentially the requirement for parliamentary consultation before deploying armed forces.  Unlike foundational conventions, they cannot be easily replaced with legal rules without undermining political accountability and risking greater judicial interference with the legislative process (subject of course to Article 9 of the Bill of Rights 1689).

13. The British constitution relies heavily on foundational conventions, as without them Britain could scarcely be called a democracy.  However, such dependency upon foundational conventions does not enhance political accountability.  This is because the actors whom they bind need not fear any political sanctions for disobeying them, only the remote threat of legislative reform.  A recent example of this occurred in October 2015 when the House of Lords defeated the Government’s planned changes to child tax credits.  This led to accusations from the UK Government that the Lords had breached a convention guaranteeing the primacy of the House of Commons on financial matters.  However, the House of Lords suffered no sanctions for acting the way it did.  The existence of the claimed convention is doubtful because the House of Lords have repeatedly denied that they are bound by any such rule (thus failing the second strand of Jennings’ tripartite text).  Even if such a convention existed, it is submitted that it should (along with other foundational conventions) be replaced with law.  This is because unelected peers are insulated from political sanction, thus reducing the incentive for them to feel bound by any convention.[3] 
14. Correcting this weakness in effectiveness, by replacing foundational conventions with law, would not necessarily signal the end of Britain’s continued dependency upon conventions and preference for political rather than legal forms of accountability.  Regulatory conventions would remain, and their role could in fact be enhanced.
15. Although there are comparatively fewer known regulatory conventions, it is submitted that there are potentially a significant number of what I have called ‘latent conventions’.  These are constitutional principles, such as respect for human rights and the Rule of Law, which guide parliamentarians in their actions and decisions, but which are not yet recognised as conventions because they do not fully satisfy the first strand of Jenning’s tripartite test above.  Latent conventions will only materialise into full conventions if and when attempts are made to subvert them, thus allowing them to crystallise fully into convention should MPs feel sufficiently strongly about defending them. 
16. Regulatory conventions have the added benefit of placing substantive limitations on the legislative powers of Parliament which are crucially not subject to enforcement by the courts.  This accords protection to human rights and the Rule of Law whilst preserving both parliamentary sovereignty and political mechanisms of accountability, thus facilitating complementary constitutionalism.[4]
17. There is the potential for an array of latent conventions embodying fundamental constitutional principles to crystallise fully into recognised regulatory conventions in the future.  This could be done expressly by parliamentarians in writing, thus accelerating the process of materialisation.

Enforceability: the effectiveness of constitutional conventions

18. The courts can recognise a convention, but they cannot enforce them, even when they are recorded in statute such as the Sewel Convention.  This was reaffirmed by the UK Supreme Court in R (Miller) v Secretary of State for Exiting the European Union [2017] UKSC 5.
19. Because constitutional conventions are legally unenforceable, it is submitted that compliance with them is dependent upon the threat of political sanction for failing to do so.  As noted above, those bound by foundational conventions, because they are unelected, need not necessarily fear any negative consequences for any failure to abide by them, save for the remote threat of legislative reform.  Such conventions should accordingly be replaced by law.  Those bound by regulatory conventions, because they are elected, are incentivised to abide by them through a threat of loss of office by either the House of Commons or the electorate.
20. However, because of the Government’s traditional dominance over the House of Commons, their dependency on anyone other than its own MPs and voters to stay in office is reduced.  So long as Government heeds the wishes of both, it rarely needs to fear political repercussions for non-compliance with such regulatory conventions.   
21. The Sewel convention is one arguable example of this.  The Sewel convention as it applies to Scotland reads as follows: ‘But it is recognised that the Parliament of the United Kingdom will not normally legislate with regard to devolved matters without the consent of the Scottish Parliament’ (Scotland Act 1998, s.28(8)).  There have been two instances since Brexit where this convention may have been breached:

a)      When the European Union (Notification of Withdrawal) Act 2017 was passed, the consent of the devolved bodies was not sought by the UK Government on the grounds that the legislation did not give rise to the need for a legislative consent motion.  The Act also passed despite the fact that the Scottish Parliament supported a motion opposing it.

b)     Although the consent of the devolved bodies was sought later in relation to the European Union (Withdrawal) Act 2018, the Scottish Parliament refused to give their consent. Despite this, the 2018 Act passed.

22. Irrespective of whether or not the Sewel convention had been breached in these instances, neither Parliament nor the Government has as yet suffered any serious political sanctions as a result of them.  This is in large part due to the fact that the convention governs the relationship between Parliament and the devolved bodies.  The majority at Westminster need not worry therefore about political sanctions for breaching the rule, as their position is not dependent upon the support of those devolved institutions which suffer detriment as a result. 
23. We should therefore question the suitability of constitutional conventions in regulating the relationship between the Westminster Parliament and the other devolved bodies of the UK.  It is possible that legal rather than political constraints may be preferable, especially if one wishes to keep the UK together.
24. On a related but separate point, the effectiveness of conventions is also linked to the scope of discretion they accord those that they bind.  In both of the above instances, it is arguable that the Sewel convention was not breached at all.  The wording of the convention, as contained under the Scotland Act, states that Parliament ‘will not normally legislate … without the consent of the Scottish Parliament’ (emphasis added).  The inclusion of ‘normally’, means that Parliament is not obliged either to seek the consent of the Scottish Parliament in every instance, or even to abide by any decision of the Scottish Parliament to refuse consent.  The Sewel convention accordingly grants Parliament wide discretion in seeking the consent of the Scottish Parliament.  The wider the discretion, the less effective the convention is as a constraint on power. In order to enhance their enforceability, therefore, it is submitted that all constitutional conventions should minimize the latitude they accord those that they bind.
25. In relation to the Sewel convention specifically, it is submitted that the discretion accorded to Parliament should be restricted to exceptional circumstances only.  Although Brexit is an exceptional event, its impact on the constitution is surely so monumental and far reaching that the consent of the devolved bodies is paramount.  It is submitted that a better limitation would be that of a national emergency, where expediency must take priority.

December 2018