The proposal to require 55% of Parliament to vote for a dissolution to have an early General Election has created much excitement, but that the number should be above 50%+1 is really a pretty straightfoward implication of the fixed term Parliaments concept. However, fixed term Parliaments are a bad idea. And, like the proposal to require new General Election shortly after any change in Prime Minister, they are solutions in search of a problem - there is, though, no such problem for them to solve.
Each of these ideas has come to prominence because of the same event, the Election That Never Was in October 2007. The claim is (a) that Prime Ministers can decide to call General Elections whenever they feel like it; and (b) that that is a Bad Thing. The first of these is false, and insofar as (a) has any truth to it, (b) is false.
Why does any informed person assume that Prime Ministers can call General Elections whenever they feel like it? What is their evidence for this claim? I presume the idea is that the Prime Minister is entitled to request a dissolution of the Queen, and that it doesn't say anywhere that he isn't entitled to do that at any time he feels like it? But, of course, the British constitution isn't written down anywhere as such (though there are of course relevant documents), so the fact that it doesn't say anywhere that he isn't entitled to call an Election whenever he feels like it doesn't prove much. Instead of a written constitution, what we have is precedent, as interpreted by the constitutional Monarch and the Monarch's advisors, as reflecting the evolving norms of behaviour within the key institutions and the expectations and understandings of wider society. So what do norms and precedent suggest?
They're pretty clear on most aspects of this point, I think. Here are the dates of the Elections since World War I:
- 1918
- 1922
- 1923
- 1924
- 1929
- 1931
- 1935
- 1945
- 1950
- 1951
- 1955
- 1959
- 1964
- 1966
- 1970
- 1974 (Feb)
- 1974 (Oct)
- 1979
- 1983
- 1987
- 1992
- 1997
- 2001
- 2005
- 2010
Now, the first things you will notice about this list is that virtually all of the gaps between General Election dates are four or five years. This gives us our first norm: A Prime Minister is very likely to be granted a dissolution if it is requested between four and five years into a Parliament. Technically, 1922 and 2005 were one month less than four years, but that points us to one of the interesting things about constitutional Monarchy and an unwritten constitution: we don't need to fuss about things like that - these dissolutions were manifestly granted under the first norm.
So, that leaves us with a much smaller set of dissolutions to analyse. Let's ignore 1945, since that was a long Parliament because of World War II. The ones left are
- 1923 (12 months)
- 1924 (11 months)
- 1931 (29 months)
- 1951 (20 months)
- 1955 (43 months)
- 1966 (18 months)
- 1974, February (44 months)
- 1974, October (8 months)
Why were dissolutions granted in these cases? Most are fairly straightforward. In 1924 and October 1974 no party had an overall majority, whilst in 1951 and 1966 the majorities were so tiny (in each case fewer than 10) that the government was in little better position than a minority administration. This leads us to a second norm: A Prime Minister is very likely to be granted a General Election if he does not command a working majority of the Commons and a General Election is believed likely to produce a more decisive outcome.
That just leaves us with 1923, 1931, 1955, and 1974.
Both 1923 and 1931 involve a similar principle. In 1923 Bonar Law resigned after being diagnosed with terminal throat cancer and rendered unable to speak in Parliament. Baldwin became Prime Minister. Baldwin took a fundamentally different view from Bonar Law on a key plank of policy, namely the question of protectionism. Baldwin wished to introduce protectionist tariffs, but Bonar Law had fought the previous General Election on a promise that there would be no introduction of such tariffs without a further General Election. So to secure a mandate for his key economic policy, Baldwin requested a General Election and was granted one (he lost).
In 1931, MacDonald's Labour government (which did not have a majority anyway) collapsed over the key policy issue of spending cuts. MacDonald indicated to the King that he was minded to request a dissolution. The King suggested that this was not a good idea, and that in such a time of national emergency, it might be desirable to seek an all-party national government in coalition. (Thus, in essence, the King refused the Prime Minister's request of a dissolution - the most recent occasion this is believed to have happened in Britain.) MacDonald duly followed the King's advice, splitting the Labour Party in the process. The split led MacDonald to believe that an Election was required. But even at this point a dissolution such a short time into a Parliament was contentious. The Liberals, led by David Lloyd George, said that a General Election was not justified and Lloyd George withdrew from the national government (most of the Liberals stayed, however). This was thus the most contentious dissolution granted in the 20th Century.
Nonetheless, we have a third fairly clear principle in both 1923 and 1931 (following on from earlier precedents such as the December 1910 Election), namely that A Prime Minister may be (but is not guaranteed to be) granted a dissolution if he feels that one is necessary to secure a mandate for the most important policy of his government.
That leaves us with two slightly curious cases: 1955 and 1974. Now both of these are elections well into the third year, and almost qualify under the four-to-five year norm, but perhaps not quite. Let us consider 1974 first. At this election, Heath argued that the key issue was "Who governs Britain?" The election was called during an extended miners' strike, which had led to the infamous "three-day work order". Thus we can understand the basis of the 1974 dissolution as lying in a combination of proximity to the four-year norm and the crisis of political legitimacy created by the miners' strike. It is pretty clear that granting a dissolution in this case was appropriate, even though it doesn't quite decisively fall under either the four-to-five year norm or the most-important-policy-mandate principle.
And that is, of course, one of the things about constitutional Monarchy. We are able to be pragmatic in such cases, and say that it was clearly sensible that a dissolution should be granted this time. We did not need to be bound by some document or other that said "In order to have a dissolution, either X or Y or Z must be true." Things happen that we hadn't anticipated, and the organic nature of constitutional Monarchy allows us to apply our discretion and do the Right Thing nonetheless.
So that leaves the slightly odd case of 1955. What was happening there? Well, we had nearly reached the four-to-five year principle, and we had a change of Prime Minister (Churchill resigned and was replaced by Eden) with the new Prime Minister wishing to secure his own mandate. So, should we feel that there is a new principle here - namely that when the Prime Minister changes an incoming Prime Minister is entitled to call a General Election to secure his own mandate (in which case, perhaps 1955 would sit with 1923); or was it simply that we were close enough to four years that there was a pragmatic case to grant a dissolution because of the change in Prime Minister even though a dissolution would not normally be granted for that reason (after all, the vast majority of changes in Prime Minister during governments with working majorities have not resulted in quick General Elections - vide Major, Callaghan, Douglas-Home, Churchill, Chamberlain, and Brown)? The truth is that we don't know...yet.
So the only case we have of a General Election since World War I in which one might potentially argue that the Prime Minister was granted a dissolution on the basis of his feeling like it was 1955, and even that case is anything but unambiguous.
But what of 2007? Well, first we do not know that Brown would have been granted a dissolution had he sought one. Would the Queen have decided that 1923 and 1955 suggested that when the Prime Minister changed mid-term, he should be granted a dissolution if he sought one, or would she have decided that, as in 1931, he ought to be encouraged to soldier on? We simply don't know, because it was not tested. For what it's worth, I believe that a dissolution should not have been granted in 2007 - it is fairly clear from the examples I quoted that the norm is that a change in Prime Minister mid-term does not require a General Election when there is no significant change in policy, when the government has a secure working majority, and there is still more than half the Parliament to run. To grant a dissolution in such a case would, I think, create a dangerous expectation that, in future, when Prime Ministers changed mid-term, they would similarly seek a General Election. That would, in turn, damage the constitution by making the Prime Minister quasi-Presidential, as if he were elected directly instead of being continuously dependent upon the support of MPs in Parliament.
But, even if my view had not prevailed on that occasion, it would still not have been the case that the granting of a dissolution were arbitrary. There would remain no reason to think that a Prime Minister that had already faced a General Election could call another General Election whenever he felt like it, except in the four- to five-year window.
So that seems to me to leave two points. The first is this: I suspect some readers will have found the above wearisome. They will have longed to exclaim: "What century are you from? The Monarch has no power to refuse a dissolution. Your long tale is pointless."
I have several responses to this. My first is this: you are simply declaring that the Monarch has no power to refuse. What makes you believe that? Formally she does have that power - on what basis are you asserting that if she chose to exercise it then her will would not be decisive? If you don't want us to have a constitutional Monarchy, then I suggest you argue your case and persuade us that your alternative system is better. I shall not simply grant you that we do not have a constitutional Monarchy and that the Prime Minister has untrammelled power. You're not entitled to simply assume or stipulate that the constitutional Monarchy provides no check on the Prime Minister's power to call General Elections, and proceed from there to decide how best to amend the constitution.
My next response is this: Before 2010, the previous hung Parliaments following General Elections were 1974 and 1929. In 1931 a dissolution was refused. So the last time but one there was any real issue, the constitutional Monarchy acted (as it had acted the previous time to that, in 1910, when the Monarch proposed a General Election). We do not want a constitutional Monarch intervening all the time - that would rather defy the point, for surely in a mature constitutional Monarchy politicians should understand the proper norms of the constitution and should not press them in such ways that the Monarch is forced to restrain what happens. The fact that the Monarch has not (as far as we know) influenced a dissolution decision in the UK since 1931 in no way suggests that the Monarch has no such power.
My last response to the "What century are you from?" point is to note that in other parts of the Commonwealth, the constitutional Monarch has acted much more recently. The most famous example is the removal of Whitlam as Prime Minister of Australia in 1975, replacing him with Fraser. Constitutional monarchs in other parts of the world have acted even more recently. The assertion that the constitutional monarch has no true power to act is nothing more than that - assertion.
That leaves us only with the much more limited concern that the Prime Minister has wide discretion over when to choose a Election in his final year in office. Is that really a Bad Thing? Do we want a Prime Minister to lose an Election just because the fixed date of an Election means that it ends up happening after some random negative event that plays badly for the Government? Obviously it means that the opposite is possible - he can call an Election shortly after some positive event, so that gives him a slight in-built advantage. But is that really so bad? Obvious modern Conservatives are in favour of revolutionary change, and I would no longer presume to pronounce upon what a true Conservative ought to think (or even to claim that I was one). But there are those deluded souls such as myself who believe that one should have a good reason for changing things, that if it ain't broke, don't fix it. We poor fools therefore don't mind if a government has a slight advantage, with the implication that Oppositions must demonstrate that there is a good reason to change and should not presume to compete on a "level playing field".
If the advantage conveyed by being able to choose when an Election were held, over the last year of a Parliament, were really large, then, yes, I might worry that the Monarch should curtail the Prime Minister's discretion further. But in such a case the Monarch could so curtail the granting of a dissolution, because people would understand why that was pragmatically appropriate. That (to say it again) is one of the advantages of constitutional Monarchy.
The requirement of fixed terms is a solution without a problem. We have no constitutional weakness in the UK in respect of the calling of General Elections. They are called between four and five years into a Parliament, unless a government lacks a majority, or a Prime Minister lacks a mandate for some key plank of policy (though dissolution is not then automatically granted) or there is some other pragmatically sensible reason for granting a dissolution. That is clear and obviously reasonable. The only motivation for changing is the bald assertion that the constitutional Monarchy could not enforce the norms above if a Prime Minister chose to attempt to violate them. But that bald assertion amounts to little more than the claim that we do not or should not have a constitutional Monarchy. If you believe we should not have one, argue for something else instead - and I shall stand against you. If you fear we do not have one, then the solution is simple: find something that all can accept for the Monarch to act upon, so as to re-assert in a visible manner her entitlement to act in constitutionally significant ways - and I shall stand with you. If, however, you simply declare it, then I simply declare otherwise.
General Elections can be valuable at times other than the exhaustion of a Parliamentary term, as illustrated above - when governments do not have majorities, when significant new issues of policy arise requiring comment from the Electorate, and some other pragmatically-reasonable occasions. Our current constitution, elegant and venerable, developed over 1200 years, allows for such cases. The Cameron-Clegg solution, dreamed up on the back-of-an-envelope in about three days, reflecting a confused reading of one event in 2007 and a ring-round for something popular to say during a Election campaign, does not.
It's almost embarrassing to have to say it: but for goodness' sake, in the name of common sense, constitutional propriety, 1200 years of history, the principle that if it ain't broke don't fix it, and the principle that we should not make bad constitutional changes for the sake of short-term political advantage, reject the 55% principle, reject fixed-term Parliaments, and reject the daft idea that changing Prime Ministers must result in General Elections. These are solutions in search of a problem that simply does not exist.
