In the wake of the exceedingly quick wrap up of coalition talks in Britain, I asked Professor Anna Bassi for her thoughts on the relationship between what had transpired there and formal models of coalition formation. Here is her response:
The new British coalition represents an intriguing case study of government formation, not just because the UK did not witness a coalition government since WWII, but because it seems to contradict common assumptions and – consequently- theoretical predictions in the field.
First, it is usually assumed that the formation process unfolds in three steps: first, a formateur (a designated prime-minister to be) is selected by the head of state; second, the formateur engages in negotiations with other parties to try to form a government; third, a government proposal (the list of members who are going to form the Cabinet) is voted by the legislature. Looking at the political events which occurred from the end of the general election on May 6th to the first Cabinet meeting on May 13th (see the Guardian’s blog), it seems that the order of the first two steps has been switched: first, parties negotiated to form a coalition, then, after a clear deal was sealed among Conservatives and Liberal Democrats, the Queen asked the leader of the Conservatives to form the government. The selection of the formateur becomes therefore uninfluential and redundant in the government formation process, since it is merely the ratification of what has been already agreed upon by the parties.
Second, theoretical models of government formation predict that, if parties are policy pursuing (as the negotiations between the parties suggest), parties will coalesce with policy-adjacent parties. Again, this does not seem to be the case, since the Liberal Democrats and the Labour party are center left parties, extremely close on several stances, with the former more radical and more leaning on the left than the latter. The Conservative Party is instead farther from both on the right. Thus, among the three major parties, Liberal Democrats, Tories, and Labour, a coalition between the first two seems to exclude the moderate – or median- party.
Third, if instead parties are not policy pursuing, but just office seekers, a coalition between the Liberals and the Conservative is –theoretically- not the best coalition to seek. The Liberal Democrats(57 seats) could have coalesced with Labour (258 seats), SDLP (3 seats), Plaid Cymru (3 seats), Alliance(1 seat), and either SNP or Green which would very likely agree on coalescing in the left wing government and get just the required majority (SinnFein does not take any seats, thus the quota required is 323). In this coalition, the relative weight of the Liberal Democrats would be larger, and therefore they could get away with a larger share of offices.
Taken together, the recent events in the UK offer important insights about how the coalition formation process should be analyzed as an endogenous free-style negotiation process among parties, rather than as formalized by specific rules concerning the selection of a formatuer. They also provide questions about what could happen in other parliamentary legislatures. If coalitions among extreme parties can form and survive despite the popular belief, intra-party politics may harshen, leading to possible parties’ splits in search of new coalition agreements. Take Italy. Gianfranco Fini, the former leader of the most extreme right party, and the second in chief of the Italian majority party (People of Freedom) has been trying for month to break free of Silvio Berlusconi and form a distinct faction. Shall UK’s events inspire him to find an agreement with left wing parties (like the Democratic Party and the Italy of Values)?
{ 10 comments }
Could use some research.
First, the protocols for appointing a new Prime Minister have a specific function with regard to the Crown. The Queen won’t choose a formateur until parliament has spoken because that would spell the end of the monarchy. At the same time the head of state needs to be more than symbolic. So we have this strange dance that appears to depend on politeness but is constitutionally vital and couldn’t happen any other way.
Second, there are no extremes among the big three parties; they are economically and socially moderate, far closer together than opposing parties in Italy (or perhaps anywhere). The difference is cosmetic, about ‘class’ and marketing. The type of people who vote Conservative are frequently the same people who vote Liberal Democrat – this can be seen by looking at the rural areas of Scotland (which hasn’t forgiven the Tories for Thatcher) which vote Lib Dem where the same areas in England are all Tory. Same people, same principles. The difference in party policies is the least important factor.
Third, the Liberal Democrats took 5 of 22 cabinet seats. Power is about leverage, not proportions. A coalition with regional parties would mean ceding far more power – or money in exchange for it, at a time of austerity. It was well known that the SNP (with a total of 6 MPs) would hold Labour to ransom in any progressive coalition.
Taken together, it’s unique to Britain.
“first, parties negotiated to form a coalition, then, after a clear deal was sealed among Conservatives and Liberal Democrats, the Queen asked the leader of the Conservatives to form the government.”
This is as far as I know the way it happens in Scandinavia as well, at least in Norway which I know best. Perhaps it’s a difference between monarchies and republics, where an unelected head of state should play a smaller role in the process than an elected president.
The third step, confirmation by the legislature, is only necessary in countries that have investiture.
If you’re going to try to make a general theory for this, you need to take into account additional characteristics of the political field. Everyone seems to have been concerned about the degree of legitimacy that the new government would have. Many said that if a government was formed by the two parties that came in last in the election, it would be seen as lacking legitimacy. The question of who would be prime minister posed an especially difficult problem for the legitimacy of a Lib-Lab coalition. If Gordon Brown, who was widely hated, had remained prime minister, this could have been disastrous for the coalition’s legitimacy. This is why he resigned. But this meant that he would have to be replaced by someone who had not been a party leader, and therefore had not been seen by voters as a possible prime minister, during the election campaign. This, too, would be seen as illegitimate, even though, as Brown pointed out, the British political system is parliamentary rather than presidential.
The issue of legitimacy must have been particularly worrisome for the Lib Dems, who were keen to have a referendum on electoral reform during this parliament. If they and their coalition partner were seen as having imposed an illegitimate government on the country, voters might punish them by voting against their proposed electoral reform in the referendum, as well as by voting against them in the next parliamentary elections. Therefore it was actually more in the Lib Dems’ interest to form a coalition with the Conservatives. Moreover, for reasons I don’t understand, there was strong opposition within the Labour Party for a coalition with the Lib Dems.
some further considerations here:
1. the sequence of moves in the coalition game was not determined by the appointment of a formateur but by nick clegg who said the conservatives should be given the first chance to form a government. this clearly hampered the chances of a lib-lab deal.
2. the “bargaining proposition” in coalition theory suggests that of all winning coalitions the one with the smallest number of parties will form. that is exactly what happened. negotiating with one partner is complicated enough. negotiating with two, three, or more really brings up transaction costs.
3. it is true that the policy distance between the two parties is quite large. however, one has to take into account that parties weigh office and policy gains differently. in the end, trading policy concessions for office benefits is what coalitions are all about.
I think fraac might be being a bit harsh. I read the comment about the formation process usually being a certain way as “usually being a certain way in most democracies we know that have coalition governments”, not implying anything about British constitutional process.
And I’m definitely against most sui generis arguments. Definitely UK to Britain? What about Canada — same constitutional jiggery-pokery, same separatist parties, same moderate differences between the parties.
I’d note this, however: if you assume, a la Comparative Manifesto Project figures for 2005, that the LibDems are to the right of Labour, then the coalition which was just formed is both minimal and connected. It’s not the minimum winning connected coalition, but it does at least make more sense.
Dear all, thanks for the comments, here are my thoughts:
To Fraac: every country is unique, with written rules and/or informal conventions that guide the government formation process. Nobody denies it. What formal theory tries to produce is a general theory free of idiosyncrasies and able to represent the institutions that affect the decisions of the players involved. Your argument is consistent with mine, that the standard bargaining theory, mostly bulding on the Baron-Ferejohn (1989) protocol, does not reflect the way the bargaining process unfolds in reality (in this case in Britain). About your comment on the three parties’ policy moderation, I’m a bit skeptical that the difference is only cosmetic: connectedness in foreign policy (war in Iraq and EU), for example, is object of debate…
To Geer: the concern about legitimacy is an important aspect of government formation, which is usually overlooked by the standard formal theory. I think that the main point is that formal theories of government formation usually do not analyze voters as relevant players in the coalition formation game (with the exception of Austen Smith and Banks, 1988). If voters are discarded and the game is not dynamic, legitimacy and credibility are not issues.
To Enser: let me clarify something. Clegg is definitely the pivotal player in this bargaining game, but formal models assume that players cannot change the institutional rules (the sequence of moves). Which “bargaining proposition” are you referring to? As far as I know, all the models of coalition formation (except Diermeier and Merlo, 2000, and Kalandrakis, 2003) predict the formation of a minimal winning coalition, which is a different concept from winning coalitions composed by the smaller number of parties.
To Hanretty: I agree. If Lib Dem were more moderate than Lab, the coalition would have been minimal winning and connected, and consistent with the predictions of models (both cooperative and non-cooperative) that assume that parties are both office and policy motivated. Furthermore, it would include the median party (among the three major parties), which would be much more in line with the theoretical expectations.
The British coalition experience appears to contradict theoretical assumptions because much of the coalition theory is itself constructed on the basis of the experiences of countries in Continental Europe. The British experience with coalition formation followed pattern similar to the Indian experience of government formation since 1989.
India has a similar institutional framework with regard to electoral rules and majoritarian parliamentary system.
So maybe we need to re-jig our theories to include a wider range of experiences.
I fail to see why anyone would be surprised that the queen didn’t appoint a formateur or that the proposed government wasn’t voted on by parliament.
That’s not how it works in the UK. Cabinets, whether coalition or not, are never formally voted on by parliament – they are appointed by the queen, and their dependence on a majority in parliament is real, but entirely informal – there is no formal process for removing a prime minister who refuses to resign after losing a vote of confidence, for instance.
As fraac notes, the formateur has to do with the fact that the queen is not supposed to interfere with the process. As I understand it, this is more or less what has happened in previous British hung parliament situations.
I also think that the idea that Labour is the median party is really problematic. There are many issues on which the Liberal Democrats and Tories are closer to each other than either is to Labour (civil liberties, for example). On a lot of economic issues, there is, at the very least, a wing of the Liberal Dems that embrace classical liberalism. If you compare the left wing of the Lib Dems to the right wing of the Conservatives, of course there’s a vast ideological gap. But Nick Clegg and David Cameron? Cameron comes from the left wing of his party, and Clegg from the right wing of his. Is it really so surprising they’d be able to work something out?
Comparing the situation to the prospect of neo-fascists in Italy allying with ex-Communists seems distinctly unhelpful.
As for the reversal of 1 and 2, I would argue that it has a lot to do with British constitutional norms, whereby a prime minister remains in the job until he resigns or is defeated in a confidence vote. Under these circumstances, Brown and Labour had a constituitonal right to continue in office while trying to negotiate a coalition, while Cameron and the Tories had a democratic right to try to negotiate a coalition as the party with the most seats/votes. Under those circumstances, where the incumbent can only be removed on facing a vote in the Commons, it seems obvious that the coalition negotiations are going to go on before someone is named PM.
As for whether the LibDems are to the right or left of Labour, it all depends on the issues you are looking at. On economic issues, they are the centrists. Also, on the civil liberties issues, it is Labour that is the outlier. The important thing to remember is that the LibDems have been at basically the same position in the ideological spectrum, but Labour has moved substantially to the right and the Tories slightly to the left, so what would have been obvious fifteen years ago is less so now. Given these facts, a Tory/LibDem coalition makes about as much sense as a LibDem/Labour coalition, although the history of Tory/Liberal coalitions suggests it will not do wonders for the LibDems come the next election.
I have to say, it’s hard to take seriously an analysis that starts with oversimplified assumptions. On the theoretical side, this analysis seems to assume that all political systems follow the Continental European model. On the political side, claiming the Liberal Democrats as being “to the left” of Labour reveals a rather oversimplified view of the relationship between the three major British parties. Any analysis that assumes that policy disagreements occur on a one dimensional axis is going to be ridiculously error-prone when it comes to predicting the formation of coalition governments.
Comments on this entry are closed.