Forecast of the upcoming Swedish election

I have always been interested in politics, so let me provide a forecast of the future Swedish election results. I conclude that the left-wing is likely to win the upcoming election as well as the subsequent three elections.



Background. All Swedish parties receiving more than 4% of the votes are represented in the Swedish Parliament. The Social Democrat Party has ruled the country for most of the time after the Second World War. Despite having 5-8 parties in the parliament, politics is always conducted as if there only existed two parties; the left (socialistiska blocket) and the right (borgerliga blocket). The individual parties have their own opinions, but coalitions are never formed across the divide (except a coalition during six years in the 1950s). The right-wing parties have ruled the country since 2006, after almost 80 years of left-wing rule. There is a new election in the autumn and the right-wing parties are likely to lose the election. The latest opinion poll shows 53% for the left wing, 37% for the right wing, and 10% for the nationalist party.

Demographics. First (figure 1) let us look at young (18-24 years), old (75+ years), and immigrant voters (source). These voter categories have forecasting value as demographics and immigration flows can be fairly accurately assessed well into the future. Older voters prefer the right and younger voters strongly prefer the left. Immigrants also prefer the left.

Figure 1. Voting patters among young, old, and immigrants. Source: SCB

Opinion leaders. Second (figure 2) let us look at categories of voters that are well positioned to influence others; teachers and journalists. The left has strong support among teachers and massive support among journalists (source, source). Up until the 1980s the journalists' voting patterns corresponded roughly to the voting patterns of the population. The state run television network is responsible for most of the political coverage. The left has massive support among those privileged journalists.

Figure 2. Voting patterns among opinion leaders. Source: see text

Forecast. Below (figure 3) is my forecast of the upcoming four elections compared to to the 2010 election. The base rate for the 2014 election is the recent opinion polls. This is also the base rate for all subsequent forecasts. I have assumed that the nationalist party will receive 16% in 2026 - similar to the situation in many other Western countries. For the other parties, I have started with data from the opinion polls and made adjustments based new voters having the same voting patterns as the current 18-24 year category. I have also assumed that a portion of the 75+ year category will die. Reflected in the results is also that one of the the small right-wing parties (c or kd) will receive less than 4%; the cut-off point for representation in the parliament. I have assumed that the current immigration flow will continue. I have assumed that teachers and journalists influence the voting patterns to a small degree. Even when I run different scenarios, the clear losers are going to be the right. The left is going to do slightly better, and the nationalists are going to do much better. It is very uncertain to make an absolute forecast 12 years into the future. However, the relative change between 2014 and 2026 is more easy to forecast as it is only driven by demographics and immigration trends.

The left-wing parties will be in majority all through the time period. It is however not likely that they would receive absolute majority in the 2014 election.

Figure 3. Forecast for future elections. The 2010 election is actual data. The 2014 election is a forecast based on current opinion polls. The subsequent forecasts are based on changes in demographics and immigration.

Additional comments. My forecast assumes that the right-wing parties do not take any action to change the situation. This is a fairly unrealistic assumption. The right-wing parties have two options:

• Enter a coalition with the nationalist party. The current right-wing prime minister has clearly stated that he will never enter any political negotiations with the nationalist party. It would indeed be difficult for him to retract. However, a new party leader can change the situation, but probably not until 2020. However, it is less likely that a coalition would be sufficient to gain power in the 2022 election. In any case, the nationalist party will get more bargaining leverage the larger its size. It is likely to be much larger in 2022 than in 2014.
• A broad coalition between the leading right-wing party (m) and the leading left-wing party (s), supported by some of the smaller parties. This has never been tested. Since the left will gain voters during the forecast period, this option is more likely to occur in the 2014 election (or in the 2018 election if the left wins in 2014 and fails to deliver). In any case, time is on the side of the left.

Given the extremely precarious position of the right-wing parties, it is remarkable that they do very little to change the situation. They do not criticise the skewed voting patterns of teachers and journalists. In contrast to most other countries, the left and the right have almost identical policies regarding immigration, despite immigrants voting predominantly on left-wing parties. The left and the right also have very similar policies on defence (=not needed) and education (=decentralisation is good and elitism is bad). Naturally, there are some differences between the parties, but the differences have probably never been as small as they are now (here).

There are three left-wing parties; social democrats, reformed communists, and the green party. For the first time in history we are likely to have the communists and the greens in government. In the past the two parties have just been obedient passive supporters of the social democrats. If the left wins the 2014 election they will have a fairly tough time agreeing and governing. They might lose absolute majority in 2018 and form a coalition with one or more of the right-wing parties. 

Irrespective of which side wins, companies generating jobs in Sweden will be favoured through tax breaks. Even the right-wing party has stated that there is no room to lower income taxes so that is not going to happen. Naturally, the left will push for a more progressive tax system. If you are an entrepreneur Sweden will continue to be a pretty good place, but if you are an employee the the taxes will eat you alive (31% on income up to USD 66,000, then 51%, and 56% above USD 100,000, in addition a GST/sales tax of 25%, and 31% social charges). For a person making USD 66,000 the three taxes mentioned above make up around USD 48,000.