Following the 1945 general election, the fifth Marquess of Salisbury, the (Conservative) Leader of the House of Lords, said the following during the Lords debate on the first King’s Speech of the new Parliament:
Whatever our personal views, we should frankly recognise that these proposals were put before the country at the recent General Election and the people of this country, with full knowledge of these proposals, returned the Labour Party to power. The Government may, therefore, I think, fairly claim that they have a mandate to introduce these proposals. I believe it would be constitutionally wrong, when the country has so recently expressed its view, for this House to oppose proposals which have been definitely put before the electorate.
This self-denying ordinance, which came to be known as the Salisbury convention, held that the House of Lords should not get in the way of the implementation of government manifesto commitments for which (by definition) the government had a “mandate” from the electorate.
The House of Lords, of course, was a rather different place in 1945. In particular, it was dominated by hereditary Conservative peers. Thus, as the parliamentary Joint Committee on Conventions explained in its report of 2006, the Salisbury convention “began, and continued, as a compact between the Labour and Conservative parties to deal with the relationship between a Labour Government and a House of Lords with an overwhelmingly large and hereditary Conservative Opposition”. Some took the view that following the implementation of the House of Lords Act 1999 — the effect of which was to eject most hereditary peers from the upper chamber — the constitutional reason for the convention, and hence the convention itself, fell away. However, the Joint Committee on Conventions disagreed. It concluded in its 2006 report that the convention persisted — and that it had become a convention that conditioned the relationship between the two chambers, rather than (as it had originally been) a mere compact between two political parties.
On this view, the convention continues to constrain — in a political-constitutional, rather than a legal, sense — the House of Lords’ capacity to obstruct Bills implementing government manifesto commitments. However, following the general election held on 8 June 2017, the question arises of whether the Salisbury convention applies in the context of a hung Parliament in which there is either a minority or a coalition government. At the time of writing, it seems that the Prime Minister envisages that the Conservative Party will govern as a minority administration, albeit with some support from the Democratic Unionist Party (DUP). I will therefore focus in this post on the position in relation to minority administrations, although I will argue that similar considerations arise in respect of full coalitions.
The Salisbury convention
The starting-point must be to acknowledge that the Salisbury convention is — self-evidently — not a law. As a constitutional convention, it represents a crystallisation of a prevailing political understanding as to how relevant constitutional actors — here, members of the House of Lords — should behave. The convention is therefore a reflection of whatever understanding exists at a given point in time, meaning, among other things, that it can evolve or even fall away. It is also important to bear in mind that the extent to which constitutional conventions matter — in the sense of setting real parameters upon relevant actors’ behaviour — is variable, and determining how “binding” a convention is is therefore a highly inexact science.
The extent to which relevant actors feel bound by a convention — and hence the extent of the convention’s prescriptiveness or “bite” — will turn on several matters. One of them is the weight of precedent: the longer a practice has been followed in the past, the stronger will be the expectation that it be adhered to in the future, and the harder it will be to resist conforming. But that can get us only so far. Members of the relevant political community are entitled to question whether future practice should take place on a basis that is in thrall to the past, and in doing so they can, and do, consider whether the underlying normative justification for the practice in question remains convincing in contemporary circumstances. This is reflected in the so-called Jennings test for constitutional conventions, according to which a practice can amount to a convention only if there is a good constitutional reason for it. Jennings appeared to see this as a binary matter: either the reason was or was not good enough to warrant characterising the practice as a convention. But a better approach is to view matters in less starkly binary terms, such that the extent (if any) of a convention’s prescriptiveness will turn upon the strength of the underlying normative justification. On this view, not all conventions are equal.
The underlying justification for the Salisbury convention is the democratic principle: the idea that it would be undemocratic, and therefore unconstitutional (albeit not unlawful), for the unelected House of Lords to block the implementation of government manifesto commitments. The question then arises as to what it is in particular about Bills implementing manifesto commitments that engages the democratic principle. It might, after all, be argued that it is always undemocratic for the Lords to thwart the wishes of the Commons, such that the Salisbury convention should apply to any Bill endorsed by the Commons, whether or not the Bill implements a manifesto commitment. But the convention does not apply that broadly — and that helps us to pinpoint more precisely the democratic concern that it encapsulates. The convention, in effect, acknowledges that Bills implementing manifesto commitments enjoy democratic credentials over and above Bills that are merely endorsed by the House of Commons — because manifesto Bills have a form of democratic legitimacy that can be traced back not just to the elected chamber but to the electorate itself. Recognising this aids analysis of whether the convention has any purchase in the context of a hung Parliament.
Minority governments
In a hung Parliament, there will, of course, be Bills that implement the manifesto commitments of the governing party. But are such manifesto commitments relevant for the purpose of the Salisbury convention? If the democratic principle underpinning the convention is understood in the way set out above, then the answer, at least in relation to a minority (as distinct from a coalition) government, must be “no”. As noted above, in 1945 Viscount Cranborne singled out manifesto Bills because the Government had a “mandate” — from the electorate — to introduce such legislation. Such a mandate had been demonstrated not by securing more than 50 per cent of the popular vote — an achievement that eluded even the post-war Labour Government, which obtained only 47.7 per cent of the votes cast — but by securing more than 50 per cent of the seats in the House of Commons. On this understanding, manifesto Bills can make a special claim to democratic legitimacy because they enjoy the support of a majority as refracted through the electoral system. A minority Government, however, has no such “mandate” for its manifesto commitments — suggesting that the Salisbury convention does not apply to them.
Of course, there are other ways in which the notions of “majority” and “mandate” might be understood for the purpose of the Salisbury convention. For instance, the then Leader of House of Commons, in evidence to the Joint Committee on Conventions in 2006, took the view that if a coalition government gained “the support of the democratically elected House and [was] endorsed by a motion of confidence then the programme for which they gain that endorsement should be respected by this House [of Lords]”. This logic would presumably apply to a minority government too, provided that its programme secured such endorsement. But this view is highly problematic, for it casts the criteria of “majority” and “mandate” so broadly as to render the Salisbury convention largely meaningless. Indeed, if the logic of this argument were accepted, then it would become difficult to see why any Bill that had been endorsed by the House of Commons should not fall within the convention. The whole point of the convention is that it recognises the special democratic claim that manifesto Bills are capable of making — and the point remains that the very notion of a “manifesto Bill” loses coherence if the manifesto in question did not secure majority endorsement via the electoral process.
Coalition governments and confidence-and-supply arrangements
If this analysis is accepted, then Bills implementing the manifesto commitments of the Conservative Party fall outside the Salisbury convention in the present Parliament because, by definition, the minority government’s manifesto failed to secure majority approval. But what if the Conservatives and the DUP form a formal coalition, meaning that the (coalition) government will then have a majority in the House of Commons? The Salisbury convention, at least for the most part, would remain irrelevant, because the respective manifesto commitments of the Conservative Party and the DUP would lack majority support as expressed via the electoral system. The only potential exception would relate to Bills implementing commitments contained — in sufficiently similar terms — in both parties’ manifestos. Such commitments could make a claim to majority support in the relevant sense, and could therefore engage the Salisbury convention.
Would the same logic apply to shared manifesto commitments in the context of a confidence-and-supply, as distinct from a coalition, arrangement? That depends upon precisely what we — or, rather, relevant members of the political community — consider to be the constitutional reason for the Salisbury convention. If its basis consists in the special democratic claim that can be asserted by a Bill with majority electoral support, then there is no reason why the convention should not extend to manifesto commitments that have such support because they appear — in sufficiently similar terms — in the manifestos of two or more parties that together command more than half of the seats in the House of Commons. That suggests that a commitment common to the manifestos of the Conservative Party and the DUP — or, for that matter, a commitment common to the Conservative Party and any other party with a combined electoral majority — would fall within the Salisbury convention.
However, the Salisbury convention might instead be understood as acknowledging the special democratic claim that can be made by a government that itself secured a mandate for its manifesto commitments, such that the convention bites only upon bills implementing the manifesto commitments of a government with such a mandate. On that view, shared Conservative-DUP manifesto commitments would acquire no special status for the purpose of the Salisbury convention, since the government would remain a minority Conservative administration, the DUP being no part of the administration itself under a confidence-and-supply deal.
The question thus reduces to whether the “mandate” that is relevant for the purpose of the Salisbury convention is one that attaches to the manifesto commitment in question or whether the convention is triggered only when the government has earned a mandate to implement its manifesto commitments. On the first view, shared Conservative-DUP commitments would fall within the convention (as would, say, shared Conservative-Liberal Democrat commitments) because there would be majority support for the commitment. But on the second view, the convention would have no application to a minority government’s manifesto commitments even if other parties, including a confidence-and-supply partner, had made the same commitments — because, as a government, the minority administration would have failed to earn a mandate to implement those commitments. Normative arguments can be made in support of either of these views. But, for reasons elaborated upon below, no answer can be dictated, because the governing criterion is ultimately what members of the relevant political community think.
England-only manifesto commitments
The Conservative Party has a very clear majority in England. And many of its manifesto commitments relate only to England, given that many of those commitments concern matters that are the preserve of devolved institutions outside England. Does this mean that Bills implementing England-specific manifesto commitments should be taken to fall within the Salisbury convention? The point is certainly arguable, the answer turning upon what we consider the relevant “majority” to be for the purpose of determining whether a “mandate” exists. If a government can claim a majority in the relevant sense only if it commands more than half of the seats in the House of Commons, then plainly there is no relevant majority even if the manifesto commitment relates exclusively to England and even if (as it presently does) the government commands a majority of English seats but no pan-UK majority. If, on the other hand, we conceive of the notion of a majority by reference to the territorial unit to which the manifesto commitment (and hence the Bill) relates, then the picture changes: an English majority would, on this view, suffice for Salisbury convention purposes if the Bill related to an England-only manifesto commitment.
Which of these views is preferable depends on how we view the UK’s territorial constitution and the roles of the UK Parliament and government in relation to it. In a formal sense, the UK Parliament and government remain UK institutions, and even when (as they often do) they act in ways that affect only England, they remain UK institutions. On another view, however, the UK institutions are the de facto English legislature and administration when they do things that relate only to England. The latter view is now reinforced by the “English votes for English laws” (EVEL) procedures that apply in the House of Commons. The existence of EVEL may be viewed as a formal acknowledgment of the UK Parliament’s dual role — as a UK and an English law-maker. On that view, it is at least arguable that the Conservative Party’s England-only manifesto commitments should be treated as having a relevant form of majority support so as to signify a mandate for the purpose of the Salisbury convention. (The same point goes for England and Wales-only manifesto commitments, given that the Conservative Party has a majority in England and Wales combined.)
Two final points are worth making in this regard. First, the fact that EVEL does not apply to the House of Lords is beside the point, because my analysis does not suggest that EVEL itself should be taken to constrain the House of Lords’ role in relation to England-only manifesto Bills. Rather, my point is that the existence of EVEL, viewed in broad terms, may be taken as a formal acknowledgment of the House of Commons’ de facto role as England’s territorial legislature — and hence of treating England as a territorial unit for the purpose of determining the operation of the Salisbury convention. Second, however, much of the most controversial legislation that the current Parliament — if it lasts long enough — will have to enact will not relate solely to England. I have in mind, of course, Brexit-related legislation, which will need to apply to the whole of the UK. It follows that even if the Salisbury convention is taken to apply to England-only manifesto commitments, that will be of only limited relevance to a minority UK government whose legislative programme will likely be dominated by Brexit.
An important caveat
The Salisbury convention, like all constitutional conventions, is an organic phenomenon that acquires its meaning, its status and its bite from the views that prevail within relevant sections of the political community. The views of that community might change over time. There may also be a diversity of views within the community. Indeed, those views may be so diverse as to preclude the existence of any view that can reasonably be considered to be a core or dominant view, thereby casting doubt on the existence of a true convention in the first place.
For these reasons, unduly legalistic analysis of conventions must be avoided. When it comes to legal rules, it is important to know whether a given form of conduct falls on one side of the line or another, because that will determine whether the conduct is lawful or unlawful. Tangible and predictable consequences will follow depending upon which of those two categories the conduct falls into. Conventions are a different kettle of fish altogether. Breaching a convention does not necessarily result in any form of sanction, albeit that it might visit all sorts of unpleasant political consequences upon the defaulting party. By same token, conforming to a convention does not necessarily render conduct in question prudential or politically wise. And, in any event, the dividing lines between forms of conduct that do and do not fall within constitutional conventions are often highly indistinct — not least because, as we move away from the core of the convention, there may be disagreement as to its penumbral reach. And, in contrast to laws, neither the need nor the opportunity tends to arise for the imposition upon such uncertainties of bright-line distinctions.
The upshot is that whether the House of Lords will — or, in a political-constitutional sense, should — feel able to disregard the Salisbury convention during the present Parliament is a question that cannot be determined in isolation from the views of the relevant political actors. And those views, given the unusual constitutional waters that we have just entered, may well be at only a formative stage. Ultimately, members of the House of Lords will have to reflect upon a variety matters, including not only questions of high constitutional principle but also much baser questions of politics. For instance, the view might be taken — as I think it should — that constitutional principle does not require Bills implementing Conservative Party manifesto commitments to be singled out for the sort of special treatment normally accorded under the Salisbury convention to manifesto Bills. But the view might simultaneously be taken that the House of Lords should think very carefully before pressing that point, given the risks that are inevitably invited by an unelected chamber that may be perceived by others to be throwing its weight around.
Ultimately, therefore, the Salisbury convention — and, more pertinently, the constitutional principle that animates it — will inform the House of Lords’ approach to Bills implementing manifesto commitments during a hung Parliament. And the Lords could entirely properly take the view that such commitments lack the special status that they enjoy when there is a majority government. At the end of the day, however, such evaluations of constitutional principle will form no more than one part of a much larger political calculus.
Public Law for Everyone 