The Riksdag has rejected two Prime Minister candidates: Magdalena Andersson (S) in October and Elisabeth Thand Ringqvist (C) on 3 November. Now Ulf Kristersson (M) is probing. Two of the Riksdag's four attempts are used up before a by-election becomes inevitable.
The distribution of mandates explains part of the deadlock. But mandate is not the same as political power.
The Tidö parties - the Moderates, the Christian Democrats, the Liberals and the Sweden Democrats - together have 173 seats and have thus lost their majority. The Social Democrats, the Left Party, the Center Party and the Green Party have a weak majority with 176 seats.
S and C are the Riksdag's power centers
Based on the parties' answers to the 35 questions in SVT's election compass, I have calculated each party's political power with a game-theoretic model developed together with Atahan Afsar, guest researcher at MIT. The measure, the spatial Shapley value (after economics laureate Lloyd Shapley), indicates the probability that a party is the tip of the scales on a randomly selected issue.
A party with its own majority has power 1, a party that never passes the decision has power 0, and the power of the parties adds up to 1.
The calculations are based on a simple model and limited data and should therefore be taken with a grain of salt. But the results are as follows:
1. S and C are the power centers of the Riksdag. S has 28 percent of the mandates but 35 percent of the power (that is, power index 0.35). C has 7 percent of the seats but 20 percent of the power – more than M (15 percent) and SD (12 percent), which have almost 3 and 2.5 times as many seats respectively.
The explanation is where the parties stand. S and C are closest to the Riksdag's ideological center, and parties in the middle are usually decisive. SD is furthest from the center, followed by KD and V, which have about 4 and 8 percent of power respectively. MP has 5 percent. L has the least power of all, just 2 percent. The party is small and in most substantive matters it is C or M who make the decision instead.
2. C is included in four out of five majority options. I have analyzed 14 potential government coalitions (see the table), where each coalition is assumed to agree on a common platform, which is an average of the parties' policies, weighted according to their power. Five have their own majority and thus power 1:
● The red-green S+V+C+M.
● The central government S+C+MP+L+KD.
● Two cross-bloc grand coalitions, S+M+C and S+M+L.
● A bourgeois pentacle, M+SD+C+KD+L.
C is included in all but S+M+L.
3. S+M+L gives the biggest power gain. The gain in power measures how much more power the parties get together than the sum of what they have individually. S+M+L gives 0.92, i.e. almost a doubling. A centrist government gives 0.52, the red-greens 0.47 and the grand coalition S+M+C 0.43. So adding C to S+M raises the power to 1 but lowers the power gain, since C already has a lot of power on its own.
Adding L instead, which has some power of its own, also gives power 1 but at the same time increases the power gain. For SD and KD, the so-called five-leaf clover (Tidö+C) would be clearly the best alternative in terms of power, but the Center Party has clearly said no to that.
4. Minority governments become weak - with two exceptions. The Tidö parties without C only gain power 0.37 and face C as strong opposition. Governments where S rules with C and some or all of MP and L gain power between 0.62 and 0.74 and a moderate gain in power.
Two minority options stand out. A government S+M gets power 0.88 and power gain 0.77, third highest of all fourteen coalitions, and faces only a weak opposition. M+C+KD+L gains power 0.67 and power gain 0.66. (A government S+C+MP+L does gain power 0.74, but a moderate gain in power.)
S and C are both close to the middle in the policy space but are far from unanimous
5. C faces a choice between power and politics. S and C are both close to the middle in the policy room but are far from united: they answer differently to 21 of the electoral compass' 35 questions and stand on opposite sides in 7 of them. The common policy of the five-leaf clover would, however, be further from the Center Party's own than the common policy of any of the three majority governments where C is part of with S. The five-leaf clover gives C the most decisive power, a collaboration with S gives C more of its policy.
Thand Ringqvist's centrist government was thus one of the five majority alternatives, but it fell when all parties except C voted no. It shows what the calculations don't capture: the parties' election promises, loyalties and pronounced red lines for which they might consider co-governing with. However, they show what each red line costs in power.
Whoever wants to form a strong government in today's Riksdag must get the Center Party on board - unless S and M are prepared to govern together.
