Fixed term parliaments are undesirable

I've already explained why fixed-term parliaments are unnecessary. I shall now explain why they are undesirable.

As I explained before, under our constitutional monarchy system, dissolutions are very likely to be granted before five years under two circumstances: if there has not been an election for about four years or more; or if the Prime Minister does not command a working majority of the Commons and a General Election is believed likely to produce a more decisive outcome. Dissolutions may also be granted (but are not guaranteed to be) if the Prime Minister feels that a General Election is required to secure a mandate for the most important policy of her government. Because a constitutional monarchy is very flexible, there are sundry other occasional circumstances, in which it will (at least sometimes) be appropriate to grant a dissolution, but which cannot all be specified in detail in advance.

If our constitutional monarchy were secure and robust, I believe that this would be a superior system. I believe that it can be made sufficiently secure and robust, though it is arguably not so at the moment. I shall explain more of how I suggest it be made more robust another time. For now I wish to focus on why a functioning constitutional monarchy-based dissolution system, reflecting the bases for dissolution I propose, is better than a fixed term parliament system.

First, let us understand why we have general dissolutions at all, for that is not so obviously required as is often thought. Our form of democracy involves representatives of local areas giving voice to the concerns of their constituents at the Centre. One could certainly imagine a form of democracy in which there were elections only when local representatives died or lost interest - i.e. in which there were local, but no national elections, and hence no general dissolutions.

There would be some fairly obvious drawbacks to such a system, however. In particular, we don’t only elect representatives of our local interests, but also someone to contribute to the national debate on national issues. And the views in each local area might evolve, or the views of the representative herself might evolve, such that the representatives views no longer reflected the views of her constituents. Alternatively, the representative might turn to be incompetent or malign in some way. It is therefore useful to have some opportunity to replace her before she would choose to stand down herself.

But we still haven’t quite reached an argument for general dissolutions. The requirement to be able to replace local representatives could be met by a system of petitioning for a local election - analogous to the “recall” concepts currently popular. Why would we need a general election?

The reason for having general dissolutions and hence general elections is that elections, in our system, are not merely a matter of representation of local concerns or even of local contribution to national debate, but also of selecting a government. To select and change governments it is not adequate that opinion in local areas be tested one at a time, and then the representatives pick the government. Under such a system whipping would not be sustainable and enduring political parties (as opposed to shifting parliamentary factions) would not flourish. We need general elections so that we can have parties that offer platforms to the voters, who then contribute to selecting the government. And because governments should not be able to put off, indefinitely, such a test, it is useful to have a maximum period before another general election must be held.

Now, for effective and stable rule, the governing party needs to command a majority in the legislature. If it does not, then it can of course seek agreement with other parties. But if the governing party could secure an overall majority were a general election to be held, why should it be forced to make accommodation with others? One obvious answer is: because that’s how the previous general election went, and it is desirable that elections occur at regular periods so that a governing party has the opportunity to implement a programme that might be initially unpopular. There is something in this thought, but if it taken to the extreme of favouring strict fixed term parliaments, it is self-defeating. For the case for extended periods without elections offered is that they protect the government from the verdict of the electors so as to encourage robust policymaking and stability of rule. But what if avoiding an election has the opposite effect - specifically, in the scenario in which the party that would secure an overall majority cannot have it because it is denied an election, and no other party can get an overall majority, either, and so rule becomes focused on popular decisions that sustain a coalition? When taken to the extreme of fixed term parliaments, extended periods without elections do not encourage robust, temporarily unpopular policymaking.

(Of course, under a constitutional monarchy one is not prevented from having coalitions - the objection offered is not that post-electoral coalitions are always bad. Rather it is that post-electoral coalitions are undesirable when an overall majority for one party (or a coalition standing as such) could be secured and is desired.)

This can be seen even more clearly in the unusual case in which a completely new party appears - say, because an older party collapses. This new party might never have faced the electorate, and its formation might reflect a step change in view or a new political cleavage-point (such as a war, or a constitutional dispute). It is, of course, useful for government to be protected, somewhat and temporarily, from the verdict of the electors. But it is also desirable that the government reflect the preferences of the electors. If a new party arises that simply hasn’t been tested, and there is good reason to believe that a general election based on the new party battlelines might be decisive, then it is desirable that a general election take place. If general elections are not possible for new parties, that will raise the bar to the formation of new parties - when otherwise a new party might need only enough MP supporters to secure a general election at a moment of high popularity for its new ideas, under a fixed term parliament system the new party will have to survive amongst MPs that were not elected to it all the way to the next general election.

A closely related case to the new party, and somewhat more common, will be one in which a new Prime Minister takes over with a new policy from within the governing party. In some cases, though perhaps not all, this might involve a major policy departure in which the Prime Minister either wants to carry the electors with her or wants a clear proof that she has electoral support so as to carry her party. Such cases offer an important opportunity for the electors to play a part in making significant policy decisions. Without it being possible to have general elections on such issues, there will be increased pressure for referendums, and the destructive seductions of direct democracy that attend them.

We have discussed two important specific reasons why fixed term parliaments might be undesirable - first, they prevent an overall majority being secured via an election, forcing more post-electoral coalitions and increasing the barrier to entry for new political parties; and second, they prevent the use of general elections to test the electorate’s view on significant policy departures. There is also the more general point that there may be numerous reasons that an early general election might be desirable, most of which we could not even imagine until they occurred. Under a constitutional monarchy we can handle these new cases as they arise.

Finally, I shall mention another reason that I think a good reason for our current system, but which I admit is more arguable. The most common basis for an early dissolution under our system is that more than about four years have passed since the last election. We don’t want there to be general elections every month - that would waste everyone’s time and prevent effective governing. A robust constitutional monarchy should certainly be able to refuse frivolously early dissolutions. At the same time, we just need a “reasonable amount” of time to have passed since the last general election, rather than being fussy about precise dates. Four years has proved to be about the right minimum in practice over the past century. There is no deep attachment to that length, however. One could certainly imagine having seven years as the maximum parliamentary term (say, if we all start to live much longer, or if political issues start to evolve much more slowly in the future - say, because economic cycles elongate) and five and a half years (say) being the normal minimum. It’s purely pragmatic.

Now to one of the things I like about this four-years-or-so principle: it means that the governing party has a slight advantage. It can choose to hold a general election at a moment when it is doing relatively well in the polls and avoid being forced to a general election when it happens to be doing badly - provided only that it does not hold on too long. This introduces a slight conservativeness into the constitution. It means that oppositions have extra work to do to beat governments. It means that we need a good and enduring reason to change, rather than a momentary spike/dip in popularity.

As a conservative, I make no apologies for favouring conservativeness. But I agree that it is not a universally held view. If there are good constitutional reasons for moving away from the four-years-or-thereabouts timescale or for giving a slight advantage to the governing party, we can change those without needing to introduce fixed terms. We could, for example, have a new constitutional convention that parliaments will last five years unless the government lacks an overall majority and can offer a compelling case to the monarch that a general election would produce a decisive result, or unless the monarch can be satisfied that there is a sufficient new policy offering that a general election would be appropriate to test the Public Will, or unless there is some other good reason (which would not include simply that more than four years had passed). Alternatively, we could say that dissolutions might occur at any time after four years have passed, but that the monarch will choose the appropriate moment without prime ministerial support, and that that moment will be chosen so as to be maximally fair to both government and opposition parties. And there are other alternatives, again.

Finally, and related to the above, I see no particularly deep reason why it should be specially for the Prime Minister rather than Parliament, collectively, or even any individual member (or some number thereof) to request a dissolution of the monarch. Since the monarch is not obliged to grant a dissolution and can force one on her own initiative, who proposes it should not be a matter of the first importance. So if there is a concern about prime ministerial control over dissolution requests then, fine!, let Parliament or some subset thereof make such requests. That is of no concern to me.

So, to reiterate, fixed term parliaments are undesirable, because they prevent dissolutions in a number of cases where they would be useful, working against robust majority government reflective of the Public Will, limiting the ability of the voters to participate in key policy departures, and unnecessarily restricting the constitution from dealing with unforeseen contingencies. The case for them? I’ll let their fans offer that.