The Winner-Take-All Circus in France and Britain
Their recent elections once again reveal the severe defects of the winner-take-all electoral system – which will also be used this Nov in US elections
[Welcome to DemocracySOS, a newsletter detailing the many dilemmas and tradeoffs of representative democracy in the US and around the world, as well as proposed reforms. If you like this issue, you can sign up here for more. Consider becoming a paying subscriber for as low as $5 per month to support this work. Or, if you already are a paying subscriber (thank you!), consider giving a gift subscription to a friend, colleague or family member. And now, let’s dive in.]
This year, 2024, has been called the biggest election year in the history of democracy. By December, more than half of the world’s population will have gone to the polls to elect new governments. While many media outlets and pundits have been hyperventilating over the alleged rise of populists and the far right in France and other countries, they have missed the really important stories going on in many of these elections. As the case of the UK and France shows, the electoral methods being used are often grossly undemocratic, leading to the very frustration and alienation from government that is fueling interest in voting for the populists.
Winner-take-all failure in the UK
Let’s look at the British elections. Most media outlets have hailed it as a stunning landslide victory for the Labor Party and its new prime minister Keir Starmer. Just so, Labor won 411 out of 650 seats in the House of Commons, a stunning 63% majority of seats and a doubling of its previous 31%. Its chief rival, the Conservative Party, which had been in power for 14 years, melted down and was reduced to 121 seats with only 24% of the vote. The 290 seat difference is its worst defeat in its 190 year history.
Those factoids have been trumpeted from headline to headline. But here’s what’s not being generally reported: the Labor Party won this unprecedented legislative majority with support from barely a third of British voters. Its 34% popular vote share was the smallest of any majority government in British history. In fact, Labor won 211 more seats than in the previous election yet it had fewer votes than last time. One commentator dubbed their victory a “loveless landslide,” and to round out the electoral absurdity, voter turnout was near a record low.
Conservative candidates in particular were hurt by high levels of tactical voting among millions of voters from the other parties who were willing to lend their vote to another party’s candidate if it would help to beat a Conservative candidate.
But don’t feel bad for the Conservatives. In the 2019 election under now disgraced party leader Boris Johnson, they won 56% of the seats with only 44% of the popular votes. In 2015, the Conservatives under David Cameron won a bare majority 51% of the seats with only 37% of the popular vote. In 2005, Prime Minister Tony Blair’s government won a majority with 35% of the vote, only 3% greater than the Conservatives, yet Labor won 157 more seats.
This kind of topsy turvy ping-pong disproportionality has long been a regular feature of British elections. The UK has not had a government that won a majority of the popular vote since 1931. Even noted leaders Winston Churchill and Margaret Thatcher never won a majority of the vote.
The reason is simple: the UK’s use of a single-seat, winner-take-all electoral system in which the highest vote getter wins, even if that candidate has less than a majority of the vote; combined with the presence of multiple political parties all competing to win seats in the House of Commons. In each district there are often many candidates running for a single seat. For decades, that has led to spoilers and split votes among like-minded parties and candidates, resulting in millions of “orphaned” voters who waste their votes on losing candidates.
Like in the US, Britain’s electoral method rewards parties that have concentrated voter bases which dominate in the greatest number of geographic districts. Both major parties, Labor and Conservatives, have suffered from the UK’s defective, crapshoot elections. But no one has suffered more than British voters, who are plagued by artificial majorities, phantom representation and other winner-take-all gremlins that bedevil fair representative outcomes.
Several of Britain’s minor parties also are regularly flummoxed by their broken democracy. Since 2005 the Liberal Democrats, usually the third largest party, has typically won 2 to 6 times as many votes as seats, such as in 2019 when it won 11.6% of the popular vote and only1.7% of seats, or in 2010 when it gained 23% of the vote and less than 9% of the seats. The Green Party also usually wins a much greater percentage of the votes compared to seats. In the recent election, the populist Reform UK Party won 14% of the national vote but only five – less than 1% – of the seats.
These minor parties underperform because of the randomness of the political map which has resulted in their partisan base being dispersed too thinly across too many districts. The combined Labor and Conservative vote share of 57.4% was the lowest for the two combined parties since the 1918 general election. That’s because several minor parties won broad popular support away from the majors. The votes-to-seats disproportionality was so extreme that it took 23,000 voters to elect one Labor MP, over twice as many voters to elect a Conservative or Liberal Democrat MP, and over 20 times as many Green voters and 36 times as many Reform UK voters to elect their MPs.
The UK’s extremely unrepresentative electoral system also has resulted in bad policy outcomes. The most glaring example of this is Brexit in 2016. Conservative Prime Minister David Cameron, after being elected in 2010 with a week mandate of only 36% of the popular vote that necessitated forming a coalition government with the smaller Liberal Democratic Party, suffered a crisis of electoral legitimacy. Consequently, Cameron courted public sentiment by making a promise to hold an in/out referendum on the UK’s membership in the European Union if the Conservatives were victorious in the 2015 general election. The Tories only increased their vote total by one point yet that led to an increase of 24 seats and their own outright parliamentary majority. The feckless Cameron didn’t even support Brexit, but too late because he had unleashed an evil genie that the UK is still reeling from today.
The British electoral system is like the crazy uncle in the attic that nobody talks about until he breaks out and causes another embarrassment. The incoming Labor government will also likely suffer a crisis of electoral legitimacy at some point, because its extreme parliamentary majority is not at all reflective of the sentiments of the British voting public. Why do British voters continue to put up with this ongoing electoral debacle, instead of switching to a proportional voting method similar to what most of its European neighbors use? Probably for the same reason US voters do – they are stuck to the flypaper of bad habit and custom, and self-interested incumbents usually try to kill most political reform.
A different type of winner-take-all failure in France
In France, the hysterical “march of the far right” headlines have now yielded to gasps of relief that the far right finished in a distant third place. The first place finisher was a left-wing electoral alliance, the New Popular Front (NFP), composed of the far-left France Unbowed, the Socialist Party, Green Party, Communist Party and several other small parties. In second was a centrist coalition, Ensemble, led by French President Macron’s Renaissance Party. Underperforming in third place was Marine Le Pen’s far right National Rally (RN) and its allies. These three electoral coalitions were fairly evenly matched, leading to a level of tri-polarization that the French have never really experienced.
Given France’s own winner-take-electoral system, this might have led to a similar outcome as in the UK, namely spoiler candidates and parties, split votes, enormous votes-to-seats distortions, tactical voting, political winners governing with a false mandate and a crisis of electoral legitimacy. But France throws a very French-like torsade into their plurality elections – they use a two-round runoff if no candidate wins a district majority in the first round. After the dust had cleared from the first round, only 76 races out of 577 in France’s National Assembly had been decided, leaving 501 districts subject to second rounds. Since any candidate who receives greater than 12.5% of the vote enters the second election, 306 districts headed to three-candidate runoffs (known as triangulaires) and 5 to four-way runoffs (known as?). With the far right RN finishing first after the first round with 33% of the popular vote, that frontrunner status unleashed waves of media hysteria that in the second round the populists might win control of the National Assembly.
But then a funny thing happened, courtesy of France’s winner-take-all elections. While Macron’s centrist alliance and the lead party in the leftist alliance, led by past presidential candidate Jean-Luc Mélenchon, were bitter political foes, they all decided that Le Pen’s far right alliance was even more odious. It was a classic “the enemy of my enemy must be my friend” situation. Consequently during the second round, 134 NFP candidates and 82 Ensemble candidates withdrew from their races, leaving one far right candidate and one non-far right candidate facing off against each other in most contests. And like in the UK, millions of voters from both the left and center alliances engaged in tactical voting, casting their ballots for candidates from other political parties if they thought it might help defeat the National Rally candidates.
The rest is now history. Le Pen’s candidates were outflanked in the second round by the consolidating alliances. Amidst surging voter turnout -- 66.6%, the highest since 1997 -- the “anyone but the far right” tag team worked, and contrary to the pre-election fear mongering, the National Rally and its allies finished third instead of first, with only 142 (25%) of the 577 seats in the parliament. The NFP candidates won the most seats, 188 (33%), and President Macron’s centrist alliance finished second with 161 seats (28%). The National Rally and its allies won only 104 runoffs, fewer than 1-in-4 of those they contested and way short of expectations.
That so-called “Republican front" or “cordon sanitaire,” the dynamic in which French voters temporarily place aside their partisan loyalties to prevent the far right from gaining power, once again prevailed. The big surprise was that so many were surprised by this outcome, since a similar dynamic had helped President Emmanuel Macron to defeat Marine Le Pen in two presidential runoffs, in 2017 and 2022. The Republican front also soundly defeated her father, Jean-Marie Le Pen, in the presidential elections of 2002 when Le Pen made the runoff and lost to conservative Jacques Chirac in a landslide, 82-18%. Nevertheless, the media ran with the “rise of the right” scare stories in its ceaseless bid to attract eyeballs and increase profits.
Winner-take-all lessons
Both the UK and French elections saw high levels of tactical voting. The winner-take-all, single-seat district system is especially vulnerable to high levels of targeting, district by district. Party leaders can do it, and so can cooperative and savvy voters who embrace their preferred party’s strategy. But something about the tactical voting strategy feels unsavory and manipulative to me. It inevitably leads to an imperfect reflection of what the electorate really feels, believes and wants, a distortion of the John Adams standard that a representative assembly "should be in miniature an exact portrait of the people at large."
But here’s the dilemma: if Britain and France both had used some kind of proportional representation system, the results would have been quite different. The votes-to-seats distortions would have been much reduced, resulting in dramatically different parliaments.
In the UK, Labor would have won 190 fewer seats and the Conservative Party would have won 35 more seats. The Green Party would have won 37 more seats, and the far right Reform UK party of lead Brexiter Nigel Farage would have won 87 more seats. That would have resulted in a House of Commons in which no party won anywhere near a majority of seats. That in turn would have necessitated the formation of a coalition government, most likely center-left led by the Labor Party with the Liberal Democrats and the Green Party. The lines of representation, alliance and accountability would have been clear — but would the coalition government have been stable?
In a parallel universe, within the strictures of the current winner-take-all rules, the options look quite different. The usual pattern in British politics, going back decades, is that when one major party benefits from such enormous winner-take-all distortions resulting in an overwhelming landslide, the advantaged party’s dominance will last for 10 to 15 years. Labor’s Harold Wilson, the Conservative Party’s Maggie Thatcher/John Major, Labor’s Tony Blair, Conservative’s David Cameron/Teresa May/Boris Johnson, all of these successive administrations more or less adhere to this longevity pattern. Due to the hardwired distortions of the winner-take-all electoral system, in the UK it usually takes a while for the governing party’s advantage to dissipate.
Certainly Labor’s new prime minister Keir Starmer will have to be skillful in his moderate Downsian targeting of the median British voter (as Tony Blair was before him). Meanwhile, the daunting dilemma for the Tories will be whether they should rebuild themselves as a moderate force under leaders such as Jeremy Hunt and James Cleverly, or should the party follow the alluring Pied Piper of populism and close ranks with the right-wing Nigel Farage? Cameron tried that with his Brexit referendum strategy, which he assumed would be rejected by the voters and he later regretted. If you play with winner-take-all fire, you can get burnt.
The electoral systems options for France are also fascinating to consider. If France had used some method of proportional representation, speculation on political scientist Matthew Shugart’s Fruits and Votes blog concludes that the far right National Rally would have won about 39 more seats and finished as the largest party with about 191 seats. Macron’s Ensemble and the NPF left alliance each would have likely lost some seats under PR elections:
National Rally: 191 (+39)
NPF: 162 (-18)
Ensemble: 122 (-37)
Nevertheless, the cordon sanitaire would still have prevented the National Rally from leading a coalition of parties capable of reaching a 289 seat majority in France’s National Assembly. Winning your 33% of seats doesn’t bring you anywhere near to power when the other 67% are aligned against you. PR would allow all political parties to win their fair share of seats and no more, essentially quarantining the ultimate impact of the far right. The Netherland’s Geert Wilders now understands this point, as more moderate conservative parties in the Dutch governing coalition have put him in his place, despite his party winning the most votes.
Values of democracy: authentic representation vs governable majorities
There is no perfect electoral system, but nevertheless some are better than others. Every electoral system has mortared into it certain democratic values. With proportional representation, the emphasis is on authentic and accurate representation of public opinion, providing a seat at the table of legislative power for every significant viewpoint in society roughly in proportion to their number of votes. These types of elections often require a coalition of political parties to form a governing majority, with those coalitions sometimes being unstable.
But with winner-take-all electoral systems, whether the UK’s one-round plurality elections or France’s two-round majority elections, the emphasis is on the election of a more stable governing majority, but at the cost of having the representation in the “people’s house” badly distorting the actual faces of the people.
Every country’s elections are complex, distinctive, even a bit peculiar in their rules, traditions and historical path dependencies. But to my value system, representativeness is the First Value, the sin qua non of democracy – “without this, nothing.” If the people can’t look at their representatives and see themselves reflected, the government fails the first test: legitimacy.
Stable governing majorities are also crucially important, but the experience of many other PR democracies show that this can be effectively achieved via coalition governments. In fact, Sweden has been run by coalition governments since the 1970s, and Germany too since 1961. The Netherlands was last run by a single party in 1879. The devil is in the details – particularly the number of political parties elected to the parliament, which is a function of how high or low the “victory threshold” to win a seat is. Some PR democracies have set the threshold too low, which has elected too many parties and caused unstable coalition governments. But these details can be fine-tuned, and on the whole, multi-representational coalition governments have proven their worth. And they tend to keep out of power any political party that can’t play nice with others in the sandbox.
Next up in this make-or-break year for democracy: an important impending election in another winner-take-all democracy – the United States. The electoral forces stampeding over the landscape are very volatile right now, and that instability will express itself via rules that, as we saw in the UK and France, tend to exaggerate election results for one side over the other. Small changes in opinion can result in large variability of outcomes. Up for grabs is the presidency and both chambers of the US Congress. In 2000, Republican George W. Bush beat Democrat Al Gore by winning Florida with a measly 537 votes out of 6 million cast. It wasn’t just Ralph Nader and the Green Party’s 97,488 votes that spoiled the election for Gore, so did the three tiny Socialist Party candidates on the ballot which together garnered 2988 votes.
In winner-take-all elections, when the contest is close, anything can happen. Fasten your seatbelts and get ready for a potentially calamitous ride.
Steven Hill @StevenHill1776
Hello Steven,
A few comments on your thoughtful piece about the UK and French elections and electoral systems. For the UK, it is clear, as you state, that PR would have led to a better outcome. A result in seats reflecting the actual support of the parties would have produced not only a more representative, but also more legitimate government to face the country’s profound challenges. Starmer ran a Labour campaign which took advantage of anti-Tory feeling and avoided any controversial commitments. It excited no one, leaving British cynicism toward politicians in place. With its landslide of seats with only a third of the votes (and high abstention) Labour will face a sophisticated media seeking to fill the role of an effectively absent opposition. Meanwhile Starmer will have to deal with an oversized caucus with many ambitious politicians excluded from the cabinet. In sum, a very short honeymoon.
France is a different story. After the instability that reigned during the fourth Republic in the postwar period, the second ballot system replaced PR. Such a system allows the French to express themselves – “let off steam” - and then go about voting practically, so that an actual workable government can be formed. The first ballot gives the various minor/fringe parties a legitimate place inside the system but the second ballot in presidential and national legislative elections is designed to prevent them from acting as spoilers when a government is formed. (This is achieved in regional and local elections under PR though a mechanism which gives the largest party half the seats plus one.) Since instituted in 1962, the system worked as expected in the large majority of elections.
In my visits to France, I asked my colleagues why France had not adopted proportional representation, like its European neighbors. PR can only work, I was told almost invariably, in Protestant countries where compromise is natural. Any single vote system, especially PR, they added, misses this essential “letting off steam” aspect of their two-vote system. (They tried PR once, in 1981, and immediately went back to the current system.)
I could see this cultural difference reflected in the rhetoric of the representatives from France and other Catholic southern European countries compared to those from Protestant Northern Europe when I was an observer at a meeting of the Socialist International. “Just because a proposal that fits our principles is impractical is no reason not to demand it” said the former, leaving the latter shaking their heads.
Similarly, while the Dutch rejection of the proposed EU constitution in the 2005 referendum three days earlier was an assertion that the EU was moving too fast toward a superstate that could hurt Dutch national interests, in France, the NON won 55% with arguments that often sounded more pro-Europe than the OUI. Thousands of posters and hundreds of op-eds proclaimed: “Oui à l’Union; non à la Constitution.”
The NON won despite the fact that the Chirac government, along with business, the left-wing national papers, Le Monde and Libération as well as the majority among the Socialists, trade unionists, environmentalists, and feminists were in favor. Elsewhere, the proponents of the OUI would and should have dominated the airwaves. But not in France, where the electronic media gave equal time to both sides so as not to be accused of being biased in favor of the establishment. I listened every evening as NON’s defenders, few of which had any position in government or expectation of having one, placed their establishment opponents on the defensive, having to respond to sweeping condemnations of a complex document of 448 articles. My sense was that had they been given a second chance, a majority would have voted in favor of the flawed constitution as superior to the status quo once having been able to voice their concerns about its flaws.
The Spring 2005 debate was “déjà vu all over again”, bringing me back to 2002 as I watched the televised presidential election debates from my Grenoble apartment. Every night, one of the 15 candidates was courteously interviewed. Never were any of the extreme left/fringe candidates asked about how their plans differed from the discredited state-socialists of Eastern Europe.
In the first round, these various fringe candidates took enough votes from Socialist Party candidate Lionel Jospin so that the far-right candidate, Jean-Marie Le Pen, placed second after Jacques Chirac. I then watched the Grenoble students, who had been passive in the first round, mobilize for the second ballot to easily elect a right-wing presidential candidate to keep a further-right one out.
The recent parliamentary election took place in a France more divided than it was then. Had it taken place under PR, the result would have reflected voters’ overall preferences as far as their preferred party is concerned. But only in a two-party system like in the US, or a two party coalition system like in Sweden would their vote for their preferred party (A) at the same time constitute a vote against the party (B) that they want to keep out of power. In the case of France today, for many voters, as we know, it is the effect of their vote on B (LePen’s RN) that is the most salient. That’s what the second vote allows.
No doubt RN supporters were disappointed with the second vote outcome, yet the party – or any other party – did not question its legitimacy.
The fact that electoral alliances were formed for the second ballot this time may mean that one can envisage a future transformation of France’s political culture into one like in Scandinavia, but given that these were negative alliances to keep a party out more than anything else, one would have a hard time convincing informed French observers that it is worth the risk that PR would entail.
It won’t be easy to arrive at a compromise to form a government this time, given the tendency of French politicians to grandstand publicly based on principle (even while negotiating behind the scenes). But had the election been fought under single vote PR, it would be that much harder.