Gamson, revised

A “law” of political science–Gamson’s law–states that parties in (post-election) coalitions divide the cabinet posts in proportion to each party’s contribution to the coalition’s legislative basis.

In most cases, these proportions would closely match their proportions of the vote, because most cases where coalitions are formed following elections use proportional-representation electoral systems. But what about the current three-way UK campaign, where we could see a coalition including a party that was badly punished by the electoral system?

In today’s Times:

Senior Lib Dem sources have revealed that if the party secures a high share of the vote in the election, it will demand equal status in any coalition.

Most current projections have Liberal Democrats in second place in votes, with around 26-29%, yet only 90-100 seats (i.e. 15% or so). Thus a revision of Gamson, based on voting shares (if not parity), would be sensible. But could Clegg bargain successfully for it?

A later Times item has Clegg suggesting he might even be PM if the coalition is with Labour. Various other indications in recent days have Clegg tilting ever so slightly towards the Tories in a post-election deal, which is a concession to polling reality–Conservatives inching ahead, Labour lagging, LibDems holding their own.

If the bargain is between Conservative and LibDem, on something like current polling, a “confidence and supply” agreement is probably more likely than a formal coalition. In that case, Clegg’s party has none of the cabinet seats (and maybe no pledge on electoral reform), but some policy influence.

It all remains fluid!

12 thoughts on “Gamson, revised

  1. Is this so-called Gamson’s law really reliable? In Norway, all coalition governments past and present has had the smaller parties overrepresented and the large one with less cabinet seats than their share of MPs or votes. The most extreme case was the second Bondevik government (2001-2005) where the Liberal Party had three seats in the government and two in the parliament…

  2. The new Tasmanian Labor-Green coalition, formed after the election, has 8 Labor ministers for 10 Labor MHAs and 2 Green ministers for 5 Green MHAs. It is admittedly not a full coalition in the sense that the Green ministers can remove themselves from Cabinet responsibility on particular issues.

    Those numbers, incidentally, show why it was such a foolish move in 1998 for the Liberals and Labor to collude on shrinking the size of the house of assembly to try and exclude the Green from parliament. Drawing a cabinet of 10 from a legislative majority of 15 in a house of 25 is madness.

    Tasmania has a tradition of using the federal single-member electorates as state multi-member electorates. The state has always had 5 MHRs only because the federal constitution guarantees an original state 5 MHRs. The obvious solution is to abolish the legislative council and elect 9 MHAs from each federal seat.

    Politicians advocating fewer politicians are invariably advocating fewer opponents, not smaller government.

  3. Seats, not votes, are the hard currency in bargaining over minister posts in coalition governments. Clegg likely has an interest in talking about votes, but his opponent will go for a seats interpretation. My guess is that in such a situation, Clegg will have to adjust. Remember, also, that UK ministers continue as members of Parliament after they take office, which means that the parliamentary party groups (seats) will be in focus.

  4. Have there been many situations where a larger party (in seats) supported a smaller party to take power-maybe India’s governemnt after the 1996 election? The United Front that actually took power was made up of a number of smaller parties, supported by the second largest party in the parliament, Congress Party, from outside the government.

    Maybe the abortive Irish free State government in 1927 that would have been led by the Labour Party, in coalition with a small right-wing party, with the support of Fianna Fail in the Dail?

  5. “Have there been many situations where a larger party (in seats) supported a smaller party to take power-maybe India’s governemnt after the 1996 election?”

    This happened twice in the UK, in 1918 and 1931 the Conservatives won a majority of seats (actually by a landslide), but supported the governments of non-Tory PMs, Lloyd George in the first instance and MacDonald in the second.

    However, for various reasons these were exceptional circumstances, and perhaps if the Tories hadn’t been campaigning for a “National” government during those elections they wouldn’t have won so many seats. At least their senior politicians thought they needed the non-Tory figurehead.

  6. Ah forgotten about those, Ed!

    If PR isn’t going to emerge from this election (I’d be surprised if it did, unfortunately) will we see instead Indian style pre-election pacts emerge in the UK?

  7. One of many reasons the Liberal and National Parties lost the 2006 Queensland State election was that the Liberal leader, Dr Bruce Flegg, flubbed a question on the hustings about whether he or the Nationals leader, Lawrence Springborg, would be Premier if their coalition won a majority.

    My understanding is that the Lib/ Nat agreement (version 2.0 at that time) specified that the leader of the coalition party with more seats would get Premier/ Opposition Leader and the other party leader would get Deputy, which sounds sensible. If they tied in seats, the party with more first-preference votes got the big one. (Throughout the 1970s, under Coalition version 1.0, the Liberals regularly polled more votes but the Nationals won more seats, thanks to a very farmer-weighted set of electoral boundaries). However, Flegg was caught in the headlights and couldn’t answer. He didn’t last long as leader.

    Nor did Coalition 2.0 last much longer. Unlike Coalition 1.0, which ended in an acrimonious split in 1983, the second incarnation led to the two parties legally merging into one, the Liberal National Party. One of the main reasons stated was to avoid losing seats through split votes in three-cornered contests (Queensland has optional preferences in State elections). I have never quite worked out why a Liberal supporter who won’t give National candidates a second preference because they are hayseed rednecks, or a National voter who won’t give Liberal candidates a second preference because they are inner-city trendies, will suddenly change and give them a first preference just because they’re now the only “LNP” candidate on the ballot-paper… but maybe I’m missing something. Anyway, by all accounts the LNP hybrid is succeeding marvelously.

  8. ‘Tasmanian Greens leader Nick McKim and his colleague Cassy O’Connor have cabinet-level portfolios courtesy of assisting Labor to form a government in Tasmania. No big shocks there — such trade-offs are a function of minority governments. The surprising news is these politicians also have opposition portfolios. McKim will be shadow spokesman for innovation, science and technology; Attorney-General and justice; and economic development, sport and recreation. O’Connor will shadow the portfolios of environment, parks and heritage; animal welfare; and the arts. Fortunately other members of the Green team are by-and-large restricted to being professional oppositionists — except when called upon to oppose the two Greens in cabinet. Yes that’s right, oppose their Green colleagues, one of whom is the party leader…’

    Katharine Murphy, “Tassie Greens wear two heads, er, hats,” National Times (6 May 2010).

    Does sound a bit odd, but I suppose the American Republic survived Secretary Mineta and Robert Gates (the latter, admittedly, hadn’t previously represented the opposition party in Congress), just as South Australia and the ACT have survived Karlene Maywald and Michael Moore respectively.

  9. … Although I suppose Maywald and Moore didn’t have party colleagues sitting on the Opposition benches at the same time.

  10. In 1886-92, 1895-1905, 1916-22, and 1931-40, the Conservatives in the UK formed governments due to splits in the Liberal or also (in the last period), the Labour parties, with dissident Liberal or Labour politicians (except during the first period) becoming cabinet ministers in these governments.

    Historians have termed the Conservative supporting sections of the Liberal and Labour parties during this period “Liberal Unionist”, “Coalition Liberal”, “National Liberal”, and “National Labour”. Except for Lloyd George, these groups tended to eventually merge with the Conservative Party (whose name is technically “Conservative and Unionist”). But the Liberal groupings took a long time to separate from the Liberal Party, and still claimed to be Liberal MPs selected from Liberal associations, even after years sitting on the Tory benches or even opposing Liberal governments.

  11. As far as I know, Gansom’s Law does not have any convincing micro-foundations yet. But it seems to me having little to do with vote shares.
    I guess if whoever wins, say, Tory, gives Lib Dem too many cabinet positions, it will create incentives for ambitious Tory MPs to defect. I don’t know if it’s ideologically plausible. But if I were them, I’ll definately make that threat.

  12. Re Queensland – I have sometimes wondered if it would be rational for the LNP to re-split but the Nationals to let the Liberals have Premier/ Opposition Leader even if the Nats get more seats. The Nats may get more first preference votes but seem to be on the nose to (my guess) 55% of the electorate. Qld has changed a lot since 1986 (when the Nats last won by a solid margin) with a lot of migration from NSW and Victoria swelling the South-eastern corner around Brisbane/ Gold Coast. Even where the new arrivals are conservative-leaning, they are accustomed to Liberal premiers and prime ministers and may find the Nationals rather alarmingly rural/ conservative for their taste. (At the same time, Labor in Qld is urban/ conservative enough to catch their votes. Former Premier Peter Beattie – currently in LA as Qld’s trade commissioner – and Wayne Goss were very small-c conservative leaders, and Kevin Rudd comes from the same tradition).

    Put another way, it could be entirely rational for a larger coalition partner to let its smaller partner have the crown if the latter has a better shot as the Condorcet-preferred candidate…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.