By Dylan Difford. Dylan is a researcher specialising in elections and voting systems. The full report on which this blog is based can be found here.

Pay only a brief glance at most of western Europe – where Proportional Representation (PR) has been the norm for democratic elections for the entire post-war period – and you’ll notice that, by-and-large, it is hardly a sea of political chaos and crises; certainly not more so than the UK and France, the continent’s only two non-PR-using democracies. And yet, opponents of electoral reform will often suggest that voting systems are a choice between fair representation and stability, presenting them as mutually exclusive. 

However, by comparing the record of 17 established parliamentary democracies – including users of both proportional and non-proportional voting systems – against ten measures of parliamentary and governmental stability across the fifty years from the start of 1974 to the end 2023, we can comfortably put this all-too-common argument to rest. 

Stability and PR

Below are the average results among both PR-using countries and non-PR users of our ten main indices of political stability – measuring everything from the term completion rates of parliaments, through to the turnover of ministers, down to the amount of time a party spends in government. On eight of the ten measures, proportional democracies actually on average outperform those that use majoritarian systems.

At the individual country comparison level, a PR user tops nine of the measures, with the time taken to form a government after an election being the only exception to the rule. Extending the analysis to 30 ‘medal spots’, 77% are taken by those that use fairer voting systems, a landslide majority if ever there was one. 

Of course, on many of the measures, the difference between categories of voting system isn’t particularly big. But, given that the charge of opponents of electoral reform is that PR is inherently destabilising, even no correlation would have been enough to disprove their favourite argument.

The idea you are sacrificing stability for superior representation is not substantiated by the evidence

The fact that PR-using countries are actually able to outperform those that use majoritarian voting systems undeniably shows that instability is far from an innate consequence of more proportional election results. In short, the idea you are sacrificing stability for superior representation is not substantiated by the evidence.

Even on the two measures where majoritarian countries come out on top, there are question marks. Governments may be formed more quickly, but that is largely a cultural choice – experience in the UK at the national and devolved levels shows coalitions can be formed quickly when there is the political will to do so. And, while changes in the government’s partisan composition might be less frequent, the changes in PR countries’ governments are often partial, with parties often spending longer in office, leading to greater ideological stability between governments. 

Early Elections

Early elections are one of simplest signifiers of political instability – often the result of governmental and parliamentary crises, with a high level of unnecessary risk if not. It has sometimes been suggested that proportional elections, with their typically higher levels of multi-partyism, necessitate frequent snap elections. The data, however, tells a different story.

The top five, including the three with no off-schedule elections (Norway, Sweden and Switzerland), are all users of PR, with most other proportional countries outperforming First Past the Post (FPTP) users Britain and Canada, the latter having held nine elections more than a year ahead of schedule in the past fifty years. Many proportional democracies, far from being riddled with the instability of early elections, are actually less likely to return to the polls before time.  

A more accurate explanation of the variation in election frequency can be found in certain institutional and cultural differences. The one thing the four weakest performers have in common is that prime ministers have significant discretion over the timing of an election, unlike in Norway or Switzerland. 

A third of British elections in the last fifty years have been caused by governments unable to deal with small or no majorities

Reactions to events such as cabinet disputes or governments lacking a majority also vary hugely between countries. A third of British elections in the last fifty years have been caused by governments unable to deal with small or no majorities, with such issues much rarer and reaction more measured in countries like Germany, Luxembourg or Sweden.

Prime Ministers

Moving to prime ministerial turnover, we again see significant variation within both categories of voting system, but with comfortably the longest average tenures coming in countries using PR.

Italy’s revolving door for the top job has long been a favourite example of FPTP defenders, even Italy has only semi-PR voting systems for the last thirty years. But Italy is an outlier and far from representative, with situations like Germany, the Netherlands and Luxembourg, where it is the norm for prime ministers to serve for the best part of a decade, clearly showing that PR is far from an agent of chaos. 

Once again, the degree of variation suggests the absence of a causal link, with differences in political culture – particularly party behaviour – key to explaining the gap between the long-servers and the short-termers. The fact that Denmark, a country with one of the most multi-party systems in Europe and where minority government is the overwhelming norm, sees longer serving prime ministers than FPTP users Britain and Canada just underlines how discordant the received wisdom is with reality on this issue. 

Governments

But probably the greatest measure of overall political stability is government durability – moving beyond just the leader of the government to incorporate how long each cabinet, often referred to as a ‘ministry’ in Britain (e.g., Blair III, Brown I, Cameron I, Cameron II, May I, etc.), lasted relative to how long it could have lasted. This takes into account the frequency of the main three change events that occur in a political system: changes in prime minister, changes in the partisan composition of governments and elections. A reshuffle of ministers from the same party or parties alone does not create a new cabinet.

At this point, seeing countries like Germany, Luxembourg and Switzerland towards the top is hardly surprising; nor is the sheer range between countries or the fact that our non-PR-users are fairly average in terms of stability. But then this is the pattern we see again and again – countries that use PR are clearly capable of the highest levels of political stability, more so than those that use majoritarian systems, reinforcing that PR’s association by its opponents with instigating instability is utterly unfounded.

The UK’s score might seem low, but our fondness for snap elections and increasingly frequent mid-term changes of prime minister mean that cabinets rarely sit for as long as they could – indeed, only two of our last six have even served a quarter of the time they could have. FPTP is very clearly not imbuing British politics with the same level of stability that PR has ‘given’ Luxembourg (where only one of the last six cabinets has not seen out its maximum term).

UK cabinet ministers served in one post for an average of just 2.1 years over the last half-century, well below the 3.1-year average in PR-using countries

Britain’s relatively weak governmental stability is even more apparent if you look below the surface, with UK cabinet ministers serving in one post for an average of just 2.1 years over the last half-century, well below the 3.1-year average in PR-using countries. And our fifty-year average hides a quickening downward trend, with the average minister appointed since the last election remaining in one post for just eight months – contributing to what the Institute for Government has called ‘chronic short-termism’ in UK policy-making.

Case Study: New Zealand

While countries using PR have invariably topped the tables above, what stands out most is the sheer variation between them, even those that use similar voting systems. A lot of this is down to differences in political culture and wider institutions. To have the best chance of discerning any definitive influence of voting systems, we need to turn to New Zealand, who switched from FPTP to Mixed-Member PR for the 1996 election, offering a near-perfect case of one political culture, two voting systems.

What’s interesting is how little difference there is before and after the introduction of PR. Most measures have remained almost identical – the only real shifters being that governments do take a little while longer to get formed (though still before the first meeting of parliament), but prime ministers and parties now actually serve longer in office. Even some ‘victories’ for the pre-reform era are somewhat illusory – cabinet durability does appear to have fallen, but when taking out the transitional first PR parliament, the post-1996 durability rate actually rises to 95%.

As ever with New Zealand’s electoral reform – the advantages of PR have materialised, the claimed disadvantages have not. 

Conclusions

On practically any measure of political stability, the stablest established democracies are users of PR, with nearly all proportional countries reaching what would be viewed as a reasonable level of stability. Whichever way you cut it, it is impossible to draw a causal link between fairer election results and higher levels of political instability, particularly with the absence of any adverse consequences since New Zealand’s electoral reform. The variation between countries that does undeniably exist is so much better explained by cultural or other institutional differences.

Ultimately, it is not credible to suggest that the UK would inherently become less politically stable if it adopted a PR voting system.  

The full report, including methodological notes, a deeper analysis of each measure of political stability and additional commentary, is available here.

Image credit: Element5 Digital