This is a guest post written by Ask Foldspang Neve.
Yesterday’s parliamentary elections in Denmark had what seem to some observers a paradoxical outcome: the incumbent Social Democrats increased their share of the vote by 1.5 percentage-points to 26.3%, becoming the largest of the nine parties represented in the Danish Folketing. However, they still lost power. That became clear as the parties willing to support a ‘red’ government led by the Social Democrats couldn’t muster a majority. Helle Thorning Schmidt resigned as Prime Minister and as party chairman immediately.
In Denmark’s multiparty system, outright majorities for one party are extremely rare––the last occurred in 1890, before the introduction of both female suffrage and parliamentarism. Governments are not necessarily coalitions, although every government since 1982 has been so. Coalition or not, however, they usually rely on the parliamentary support of a ‘friendly opposition’, i.e., parties that will reliably support the government in key votes such as on the budget, but are not part of the government’s official program (agreed upon among coalition partners and published somewhat like the Queen’s Speech).
Most often, the support of the friendly opposition is secure enough that even if such supporting parties do not take part in every law the government passes (private member’s bills are extremely rare), they will still make sure that the government stands, although often in return for some kind of policy compensation.
Usually, the logic ensuring that the government is safe––even when it makes deals with the ‘real’ opposition that are unpopular among its supporting parties ––is that supporting parties lie further to the extremes of the political spectrum than governmental parties. Thus, even a ‘promiscuous’ government is often more attractive to a supporting party than the alternative of getting a government led by a party whose preferences lie even further away from oneself.
The obvious situation where this might break down is if there is more than one dimension of salience represented in parliament. With 10 parties running for election, a low 2% threshold for entering, and a voting system that ensures a very high level of proportionality between votes and seats, Denmark has the institutional setup for expressing just that.
A recent historical example occurred in the 1980s, when, during the Cold War, the Conservative-led government relied on the support of the anti-militaristic Social Liberal Party for support in budget votes. The Social Liberal Party, however, voted together with the left on defense issues, partially in defiance of Ronald Reagan. Thus, to remain in government, the Liberal Party foreign minister was forced to negotiate international treaties and policies that went against his own and the government’s will; the alternative would have been to call fresh elections, which the prime minister was unwilling to do over the issue.
From the mid-1990s onwards, the two salient dimensions have been redistributive preferences and social attitudes––especially concerning migration and religion, but also crime (gay rights are only a very minor issue, and abortion is absent from the public discussion). European cooperation might mark a third dimension.
In this year’s elections, 10 parties were running, 5 in each ‘bloc’ supporting a different candidate for prime minister. There are 179 seats in Parliament, of which 175 are elected in mainland Denmark, 2 on the Faroe Islands and 2 in Greenland. The latter 4 mandates are usually part of supporting a government in key votes if needed (according to their political alignment), but otherwise usually remain neutral. Thus, 90 mandates are needed for a majority. Yesterday’s results in mandates for each party and for the blocs are listed below.
However, the Social-Liberal Party has traditionally been a ‘swing party’, and has participated in governments in both blocs. This is not least due to the party’s economic preferences, which align more closely with those of the Liberal Party and Conservative Party. On social issues, they are aligning mostly with the left at the same time as the Social Democrats have moved to vote with the right.
Similarly, the Danish People’s Party have expressed that in order for them to support a Liberal-led government, the Liberal pre-election promise of freezing public expenditure must be reneged on: the Danish People’s Party’s economic preferences are to the left of the Liberal Party and maybe more closely aligned with the Social Democratic Party, from whom it has managed to take many voters. Thus, if parties votes according to this preference––without paying respect to supporting a government––we could list the mandates that could be mustered for each side on multiple dimensions. I have arranged them below.
Thus, if parties paid no respect to office, we should expect to see agreements securing higher levels of redistribution, lower levels of immigration, and more European integration. By some measures, this has actually been the case in previous assemblies. Anders Fogh Rasmussen’s liberal governments saw increases in tax revenue as share of GDP (in a period when GDP growth was high) while his government was supported by the more redistribution-friendly Danish People’s Party. The Social Democrats are largely seen as having moved to the right on immigration, and Ms. Schmidt’s government made getting asylum more difficult, despite the protests of Social Liberal backbenchers and the supporting parties on the left.
Finally, while Euro-skepticism has been growing, there is still a stable consensus about supporting the European Union and to resist calls such as those heard from Tory backbenchers for scaling back or even quitting the union.
It shows why the Danish People’s Party might be doing so well: it is good at capturing voters who want more redistribution, but less immigration. It often seems isolated on European issues, which, however, is a dimension that is often less salient than the first and second dimensions. In this sense, the current election might not have changed much. However, governments matter: office spoils are important, and only governments can effectively initiate legislation in Denmark. Moreover, most parties obviously don’t fit well into a dichotomous description of any dimension of politics, and the above 3x2 matrix does not even allow for centrist positions.
Nonetheless, the above data gives an impression of multiple dimensions at work in practice in a multiparty system––and of what to expect from the ongoing negotiations to form a government in Denmark.