The Fragmentation of the Electorate in the West

Last month’s General Election in The Netherlands was won by a party with less than 17% of the national vote. Dutch voters, politicians and pundits are now discussing what this means for a governing coalition which is likely to take a number of months complete. For myself, the election was yet another data point showing the fragmentation of the electorate in the West is the key trend over the last 40 years. As yet, there is no sign of this trend coming to an end and what the future holds I cannot tell.

General Election Trends in 6 Countries over 60 years

Take a look at the 6 charts below. All show the national vote shares in each election since 1965 for the two main parties (for which I will use the term Duopoly as a collective noun in this article) in those countries plus the share for All Other parties combined. For the countries who did not have an election in 2025, I show the average of voting intention polls in October 2025.

The 6 countries have three things in common. Can you see what they are?

More and more people voting for All Others.
The trend to All Others exists irrespective of election system used.
There is no end in sight for this trend.

These are the three points I want to discuss in this article. There are two other points the 6 countries have in common which are the reasons why I decided to use them in this article.

These elections are for a parliament from which the head of government emerges. In other words, there is no elected figure with more powers than the head of government as happens in the presidential elections of France and the USA.
The frequency of elections is similar in all countries with average parliamentary lifetimes of just under 4 years. Australian elections must take place every 3 years which is why they are more frequent.

So what conclusions do I draw from the commonalities I’ve highlighted?

The Fragmentation of the Electorate is a 40 Year Steady Trend

I will explore the trends in each country separately at the end but here is the average vote share of All Other parties for each country for all elections in 10 year timeframes. The combined columns is the average across the 6 countries. The main reason I don’t go back further is that in many countries, the lifespan of any Duopoly of the two main parties rarely goes back for more than 60 years.

For the first 20 years, the Duopolies attracted an average of 78% of voters with All Others on 22%. However, from the late 80’s this started to change and by the most recent period 2016 to 2025, All Others has more than doubled leaving them neck and neck with the Duopolies on 50% each. The 40 year growth in All Others averages 7% per decade with no sign of any slowdown. If anything, the trend has accelerated over the last decade

Electoral Reform is Distraction Therapy

I am aware of much talk on social media about electoral reform in both the UK & Australia recently. It would not surprise me if similar talk is taking place in the other four countries. Advocates of electoral reform have pointed to the fragmenting electorate as justification but I come to a different conclusion. I see a similar trend in all 6 countries, each of which has a distinctly different electoral system, namely –

UK – First Past the Post (FPTP) in single member seats.
Germany – Half elected with FPTP in single member seats, half elected with Regional Proportional Representation (PR) using either open or closed lists subject to a national threshold being reached.
Australia – Ranked Choice Voting known as AV (Alternative Vote) in single member seats.
Ireland – Ranked Choice Voting in multi member seats known as STV (Single Transferable Vote).
Sweden – 90% elected with Open List PR in multi member districts, 10% elected with a national Closed List PR
Netherlands – National Closed List PR

The 6 countries more or less cover the full gamut of possible electoral systems. In an ideal world, I would have liked to have included France as well with its Run-Off system over two rounds for single member seats. But France’s duopoly didn’t really establish itself until the 70s and was less durable and of course France elects an executive President which is why I don’t include them.

The point is though, if the same trend exists in all 6 countries irrespective of its election system, then surely changing the electoral system is going to have zero impact on the underlying trend to fragmentation? That doesn’t mean you can’t argue for electoral reform (see my preference here) but it is not the answer to voter fragmentation.

When will voters stop fragmenting?

I haven’t a clue.

Obviously the maximum vote share for All Others is 100% but there is still room for further fragmentation as the Netherlands shows. It is worth noting there is little sign of any new party among All Others who can command ~40% of the vote in any of the 6 countries. The ceiling these days appears to be around 33% of the vote. Connected with this point is the fact that in all 5 European countries, there is a non-duopoly party who is or was polling at over 20% who have been deemed “beyond the pale” when it comes to forming a government, namely –

UK – Reform, have led the polls since March 2025 with ~30%. However, the question of whether other parties would refuse to form a coalition with them has not yet been tested to any extent.
Germany – Alternative for Germany (AfD), recently started leading the polls with 25%+.
Sweden – Swedish Democrats (SD), currently 2nd on ~21%. They support the current government but are not part of the government.
Ireland – Sinn Fein (SF), have led the polls since March 2025 with ~23%. They came first in the 2020 general election with 24.5% of 1st preference votes but ended with fewer seats than Fianna Fail after preference redistribution.
Netherlands – no party appears capable of reaching 20% these days but Geert Wilders’ Party for Freedom (PVV) party came first with 23.5% of the vote in the 2023 general election and was part of the previous 4-party coalition government.
Australia – no non-duopoly party as yet polling over 20% on 1st preference votes but take note of a point I make in the section on Australia below.

On the evidence above, I doubt “beyond the pale” politics can be maintained unless voters stop & reverse their fragmentations. I see no sign of this happening.

**Further reading – I wrote this comment 6 years ago in answer to a question posed on the Quora site. In it, I explore reasons why fragmentation & polarisation was on the rise using Northern Ireland as an analogy. I think it still reads well and is connected with the theme of this article.

Country Reviews – Data used & Background

I finish with a review of each country. For each, I give Wikipedia links to the election results and available voting intention polls. I also point out some nuances to be aware of in each country’s politics.

United Kingdom – House of Commons

Election Results here, Voting Intentions here.

The duopoly in the UK is one of the most established worldwide. The Conservative & Labour parties have taken the top two spots in vote share at every general election since 1922. Neither party has undergone any significant renaming or merger in that timeframe.

The trend in All Others began the 1970s with the recovery of the Liberals (now Liberal Democrats), the establishment of the various Northern Ireland parties and the Welsh & Scottish Nationalists winning their first seats. From 1997, UKIP started its rise to prominence before mutating into the Brexit party & now Reform. The Green party won their first seat in 2010 and are now climbing in the polls.

The most recent election saw the lowest ever vote share for the Duopoly of just under 60% but could just about be read as a continuation of their hegemony. Since then, the polls have been tumultuous and the Duopoly now polls under 40% with Reform leading the polls since May on 30%.

Germany – Bundestag

Election Results here, Voting Intentions here.

Germany’s duopoly of Christian Democrats (CDU), which includes the Bavarian sister party Christian Social Union (CSU), and the Social Democrats (SPD) was established in the first post world war 2 election of 1949. These two topped every federal election since until 2025 when the Alternative for Germany (AfD) beat the SPD into 2nd place with 21% of the list vote. Most polls now show AfD in first place, narrowly ahead of the CDU/CSU.

However, AfD did not start the rise of All Others in the 1980s and indeed only won their first seats in the Bundestag in 2017. The 1980s rise was led by the Green party supplemented by Linke in the 1990s after the union with East Germany. Germany now has a 6-party system and came close to establishing a 7-party system in 2025 after the BSW secured 4.98% of the list vote in 2025, a fraction below the 5% threshold for parliamentary representation.

Australia – Lower House

Election Results here, Voting Intentions here.

The Australian duopoly of Labor & Liberal-National coalition began in 1946 when the Liberal party was formed. Until 1983, the National party was called the Country party. There have been no serious challengers to this duopoly with the highest 1st preference vote shares by challenger parties being the 12.3% for the Greens in 2022. The Greens vote share has exceeded 10% in 5 of the last 6 elections.

Despite the lack of serious challengers to the duopoly, Australia has the distinction of having the most predictable trend for All Others of the 6 countries I’ve looked at. If you look back at the table where I show trends by decadal timescales, the last two rows shows the mean and standard deviation for the change in vote share between elections for All Others. The standard deviation for Australia is 4% compared to 6% for Germany & Ireland and 8% for the UK & Netherlands. That means a projection of the trend into the future has more predictability in Australia than elsewhere.

The same table shows the mean change between election for All Others averaged +1.1% over the last 60 years. However, this disguises a possible acceleration. Up to 2007, the mean was +0.3% (Stdev 4.3%) but since 2007, the mean is now +3.2% (Stdev 1.8%). The apparent acceleration can be regarded as statistically significant (P-value 6%).

Ireland – Dail

Election Results here, Voting Intentions here.

The Fianna Fail & Fine Gael duopoly started in 1923 like the UK’s when the forerunner to Fine Gael split from Sinn Fein under the name Cumann Na nGaedheal. They became Fine Gael in 1933. They shared the top two places in all elections until 2011 when Labour came 2nd to Fine Gael with 19.4% of 1st preferences. The Duopoly regained the top two places in 2016 but Sinn Fein then overtook Fine Gael in the 2020 & 2024 elections.

Sinn Fein continues to be the main party leading the rise of All Others but several other parties are polling in the 3-6% range, all of which add to demise of the Duopoly. This means that unless more of these parties can increase their vote share, Sinn Fein will never be able to form a governing coalition unless either of Fianna Fail or Fine Gael choose to do so.

Sweden – Riksdag

Election Results here, Voting Intentions here.

The Social Democrats have come first in all elections since 1917 and this doesn’t look like changing soon. They continue to poll between 30-35%.

I had to stretch to mark the Moderates as part of the Swedish duopoly along with the Social Democrats. They came second to the Social Democrats in all elections between 1979 & 2018 and between 1917 & 1944 but they were displaced between 1948 & 1976 first by the Liberal Party and then by the Centre Party. In 2022, they were displaced by the Swedish Democrats with 20.4% of the votes.

Sweden had a 5-Party system since 1917 comprising of the Social Democrats, Left Party, Liberals, Centre Party and Moderates. This explains why All Others in the chart were already polling at ~40% in the 1960s. They were joined by the Christian Democrats in the 1960s, the Greens in the 1980s and the Swedish Democrats in the 2010s.

Netherlands – House of Representatives

Election Results here, Voting Intentions here.

If Sweden was a stretch to identify a duopoly, then the Netherlands was very much an elastic band taken to breaking point! Like Sweden, All Others started the 60s & 70s on ~40% but unlike Sweden, the Dutch duopoly’s elastic band has snapped.

The current Christian Democrats (CDA) came about in 1976 as a merger of three parties so what’s shown in the charts for the 1st three elections is the sum of those parties. Labour exist as a separate party but in 2022, they formed an alliance with the Greens and now compete under a joint banner.

The last time the Christian Democrats came 1st in an election was in 2006. The last time Labour topped the vote was in 1998. The duopoly’s peak came in 1986 which is the only time when All Others combined vote share of 32.1% was behind the vote shares of both the Christian Democrats (34.6%) and Labour (33.3%).

Since 2010, the election winners have been the VVD (People’s Party for Freedom & Democracy) from 2010 to 2021 with vote shares between 20% & 26%, the PVV (Party for Freedom) in 2023 with 23.5% and now D66 (Democrats 66) in 2025 with 16.9%. As my wife likes to say, the Dutch no longer have political parties, they have high school cliques! I can’t think of a better metaphor to describe the fragmentation of the electorate in the West over the last 60 years.

— Would you like to comment on this article? —-

Please do leave your comments on this X/Twitter thread or this older X/Twitter thread from August. The 1st link starts with a retweet of another thread connected with the 40-55 Channel I wrote about in July but I then use that to bring in the point of this article.

—- Want to read more about polls and polling errors? —

More articles on Elections & Forecasting can be found by clicking on the relevant tab at the top of your screen.

— Subscribe to my newsletter to receive more articles like this one! —-

If you would like to receive notifications from me of news, articles and offers relating to Polling & Elections, please click here to go to my Newsletter Subscription page and tick the Forecasting and/or Elections category and other categories that may be of interest to you. You will be able to unsubscribe at anytime.