Why Proportional Representation is better for workers

Robert Ford lays out the injustices of the first-past-the-post system, and demonstrates why trade unions should back PR

The current electoral context in Britain

In Britain, the electoral geography is bad and getting worse for Labour for a number of reasons. The latest boundary review will decrease the number of MPs in Labour areas by reallocating seats away from slow growing, Labour-leaning areas such as Wales and towards more rapidly growing, Conservative-leaning areas such as the South East. The fall of many low-turnout, erstwhile Labour seats in the ‘red wall’ has also removed a traditional Labour advantage – their seats were often won with fewer votes than in high-turnout Conservative areas.  There are also more complex strategic problems: Labour loses more seats narrowly to a third party, usually the SNP, than the Conservatives do – all those votes cast for Labour do not return more Labour MPs. Conversely, the Conservatives have more seats where they narrowly beat a third party, usually the Liberal Democrats – these Tory votes add Tory seats. Last, but not least, is the problem of safe seats: an awful lot of Labour votes are cast in seats where the incumbent Labour MP already has a large majority. These extra votes add to the local MP’s job security, but do not elect more Labour legislators. When we put these factors together, they add up to a large and growing disadvantage to Labour from the electoral system, with huge majorities in many urban core seats and narrow defeats elsewhere, leading to a high number of votes ‘wasted’ in the sense that they do not contribute to returning extra Labour MPs. 

Sir John Curtice has summarised the situation after the 2019 election as follows: “For any given performance, the electoral system is inclined to reward the Conservatives more richly than Labour.” If the vote shares at the next election were even, Curtice predicts the Conservatives would have 23 more seats than Labour. It is also much harder for Labour to get a majority: the Conservatives need a 3.5 point lead in votes to get a majority at the next election, while Labour need a 12.3 point lead to do so. Therefore, there is also a large range of outcomes where Labour would have a lead in votes cast, but Conservatives would lead in MPs returned to Westminster.

Labour’s disadvantage from the electoral system is not new, as figure 1 below, from analysis by political geographers Ron Johnston and Charles Pattie, illustrates. In most post-war elections, the electoral system favours the Conservatives. Only during the New Labour period was the voting system inclined to reward Labour more generously, and the unusual coalition of voters assembled in that period looks, at present, unlikely to be reassembled.

Figure 1: Biases in the electoral system 1950-2015 (Source: Pattie, C and Johnston, R (2015) “Electoral Bias in the UK after the 2015 General Election”)

The particular problems caused by very safe seats are illustrated in Figure 2, which shows the distribution of majorities in Conservative and Labour-held seats after the 2017 election, when the parties’ overall vote shares were closely matched. Conservative majorities tend to cluster in the 20-40% range – safe but not landslide majorities. By contrast, a quarter of Labour seats have majorities over 40%, and one in eight has a majority above 50% – seats where the Labour vote is not so much counted as weighed. These huge majorities are inefficient – adding to Labour’s national vote share, but without any prospect of returning extra Labour MPs. The same problem is evident in 2019, when the ten largest majorities in percentage terms are all Labour-held seats in Liverpool, Manchester, Birmingham and London. Nineteen of the top 25 majorities in the last election occurred in Labour seats. 

Figure 2: Distribution of majorities in Labour and Conservative seats, 2017 (Source: author’s calculations)
Why the electoral system matters

In 1968, Giovanni Sartori pointed out: “[The electoral system is] the most specific manipulative instrument in politics.” What he meant by this is that the rules applied to turn votes cast into seats allocated have a powerful impact on the kind of politics countries get. This impact comes through two channels. There is a mechanical effect, which is simply the impact of applying the system’s rules to determine who gets elected based on the votes cast. But there is also a psychological effect: voters know the rules of the system, and this affects how they behave. For example, in first-past-the-post systems, voters who favour smaller parties, or locally weaker parties, may recognise that their preferred candidate is a lost cause and back a second preference candidate instead.   

The mechanical effects alone of our current electoral system are large, as illustrated in figure 3 below, which uses calculations from the Electoral Reform Society’s publication ‘Voters left Voiceless: the 2019 General Election’. The only system which returns a Conservative majority is the first-past-the-post electoral system currently used. The primary effects of shifting to any other system are to reallocate seats away from the Conservatives and towards smaller parties.

Figure 3: MPs returned from 2019 votes cast under different electoral systems (Source: Garland, J; Palese, M and Simpson, I (2020) “Voters Left Voiceless: The 2019 General Election”, Electoral Reform Society)

The impact the electoral system has on the politics of Brexit is illustrated in figure 4, where I show how seats would be allocated to parties favouring the Conservatives’ Brexit approach or favouring a different approach under different electoral systems. The only system which delivers a Commons majority for the approach to Brexit proposed by Boris Johnson is the system we currently have.

Figure 4: Conservative/UKIP/Brexit Party vs Labour and Others in 2019 General Election under different electoral systems (Source: Author calculations, based on Garland et al (2020))
The case for FPTP and why it now often fails

The traditional defence of first-past-the-post proponents is that it is not a system designed to equally represent all those who stand for election, but rather a system which delivers “strong”, “accountable” governments – strong, because it delivers legislative majorities even on a minority of votes; accountable, because it enables voters to remove one majority government and replace it with another. However, FPTP has not consistently been delivering either outcome for some time, and is unlikely to do so again in the coming election. There has been only one substantial majority government in four elections since 2005. This trend is likely to continue: territorial fragmentation has produced a large rise in third-party MPs, and geographical polarisation has produced a large decline in traditional Conservative-Labour marginals. The range of outcomes which would deliver another hung parliament at the next election is very wide indeed.

The case against FPTP

A number of arguments are made against first-past-the-post by proponents of reform. Most constituencies in an FPTP system are ‘safe’, meaning most voters do not get a meaningful choice at any election and, as parties may safely ignore them in campaigning, they do not receive much attention from local or national campaigns. Instead, elections are mostly fought in a narrow set of ‘marginal’ seats, where activists and campaign resources are focused, while the voters of other seats are marginalised and forgotten. Swing voters in marginal seats therefore have disproportionate influence, but there is no guarantee that such voters will reflect the distribution of views in the country. FPTP also raises unfair barriers to smaller parties, therefore denying voters a meaningful choice and preventing effective representation of voter preferences. Turnout and engagement with the political system tends to be lower under FPTP. Comparative research has demonstrated that proportional electoral systems are associated with a large range of progressive outcomes, including: greater representation of women and ethnic minorities; greater representation of left-wing voters who tend to concentrate in urban areas; more left-led governments; lower income inequality due to greater redistribution through welfare spending; higher trade union membership, coverage and influence; more progressive labour laws.  

In particular, it is overwhelmingly accepted by academics that FPTP has a pronounced bias towards centre-right parties. Stanford Professor Jonathan Rodden says: “In every industrialised parliamentary democracy with majoritarian electoral institutions, averaging over the post-war period, the legislature has been well to the right of the voters, and in most cases, the cabinet has been even further to the right”. This reflects the biases produced by voter geography – left-wing voters are, in most countries, more inefficiently distributed as they concentrate in a small number of strongholds (usually large cities and towns with large working-class populations), while more even support enables right-wing parties to win narrow majorities over a wider range of seats.

The right-wing advantage under FPTP also reflects voter psychology – risk-averse middle-class voters are often willing to back centre-left coalitions under PR, but when forced into a binary choice between left and right, they tend to align more with the right.

The cumulative effect of geography and psychology is large: countries with FPTP systems tend to have right-of-centre governments two-thirds of the time, even though left-wing and right-wing parties have received a similar share of the vote on average over the long run.  

Electoral systems and workers’ rights

Trade unions tend to have more influence under more proportional systems than under first-past-the-post. Looking at the International Trade Union Confederation’s Global Rights Index, all the top-rated developed democracies use PR. The UK is graded as 3 – “regular violations of rights”, alongside other majoritarian countries such as Canada and Australia.

Figure 5: Electoral Systems and trade union rights (Source: International Trade Union Confederation “Global Rights Index”)

This is one example of how it is in the interests of trade unions to have a more proportional system, because more proportional systems deliver more left-leaning governments and more progressive policy outcomes. On average, proportional democracies tend to have more government ministers from social democratic Labour-type parties (35% of ministers) compared with majoritarian/FPTP countries (23%). A more proportional system aligns with or advances many trade union goals and, in providing fair and equal representation of voters, is in itself a progressive goal.

The impact of reform: a Kiwi case study

New Zealand switched from first-past-the-post to a mixed member electoral system in 1996. Large majorities of voters ever since have said they prefer the new system, and political parties across the ideological spectrum have become steadily more supportive of it over time. The new system has delivered more ideologically diverse parliaments, with greater representation of small parties; yet the traditional “governing parties” – Labour on the centre-left and the National Party on the centre-right – have continued to lead governments, providing all prime ministers and most senior ministers. Since electoral reforms, there has been more representation of women and Maoris in Parliament, and voters express higher trust in government. One academic studying the reform, Jack Vowles, concluded “the most important expectations of New Zealand’s electoral reformers have been met”.

Electoral reform has altered the balance of power between the two largest New Zealand parties. The centre-right National Party governed for 74% of the time in the 47 years before reform was introduced; in the 26 years since, the centre-left Labour Party has governed more than half of the time. This leftward shift in the political centre of gravity has had major policy consequences. The National Party government introduced some of the most individualistic labour legislation in the world in 1991. In 1999, under PR, Labour formed a government with the left-wing Alliance Party and repealed the National Party’s trade union legislation in 2000. Since 2000, New Zealand Labour has introduced legislation on paid parental leave, equal employment opportunities, rest break and breastfeeding, and flexible working relations.

Figure 6: Net positive mentions for trade unions or workers’ rights in New Zealand National and Labour election manifestos 1946-2017 (Source: Comparative Manifestos Project)

Trade unions and workers have also benefited from electoral reform in more subtle ways, as illustrated in figure 6, which charts the net number of positive mentions of trade unions and workers’ rights in New Zealand parties’ manifestos. After the switch to a more proportional system, positive mentions of unions and workers’ rights increased dramatically and have remained elevated ever since. Even the centre-right National Party has begun to make positive mention of these issues in recent elections. Proportional representation has returned New Zealand’s workers, and their working conditions, to the heart of the political conversation.

Prof Robert Ford
Robert Ford FAcSS is professor of political science at the University of Manchester. He researches elections, public opinion and British politics and is the author (with Tim Bale, Will Jennings and Paula Surridge) of The British General Election of 2019 and (with Maria Sobolewska) of Brexitland.


  1. 1. PR Does not necessarily lead to a higher level of women being elected. They have PR and high women’s representation in Scandinavia. But that is because it is Scandinavia, where they would have a higher level of women’s representation anyway, under any system. The proportion of women being elected is, quite rightly, on the rise across mature democracies, from a low base. Therefore, if you take any event (whether it is the introduction of PR in New Zealand, a new television programme, or anything else), a before and after comparison will show more women elected after. That does not necessarily mean that PR was the cause.
    2. PR discriminates against small parties, as they are often locally based. The local issues then get drowned out by other parties and other issues. Could the Save Kidderminster Hospital group have got elected under PR? They would have needed the votes of people who were not very near to Kidderminster. That is unlikely to have happened.
    3. The elephant in the selection meeting. Who chooses the candidates? There is currently an expectation that the local branch (CLP, Conservative Association etc) of a party will get to choose their candidate. This is flawed, as there is increasing pressure and outright manipulation by the parties to get the outcome that the leadership desires. Nevertheless, the expectation is there. A candidate and an MP is expected to “work their patch”, and my face a challenge if they do not. If you have centralised candidates on a list system, they will almost certainly be chosen at the centre. They will put leadership clones at the top and anyone that they do not like way down the list past the chance of getting elected. Remember the Euro elections in 1999? All local accountability will be gone. Sure, the proportions elected will be more “accurate”, but they will be decided by a few dozen middle ranking apparatchiks in the big parties, not by the people.

  2. Once again, we hear the chorus of the fainthearted and the unable to see the bleeding obvious. For the hard-of-understanding, you first have to win under the present system! The point is exactly that Labour cannot “slip in when voters aren’t looking” – it has to assemble a winning coalition in the here and now. If it wants a second term, then it has to fulfill its material promises in the first four years. It has to win over voters by addressing their needs. If that’s “too populist” for some, then tough.

  3. One solution to the second point you brought up is that the voters choose at the same time (on tne day of election) BOTH the party they wish AND indicate the order of the party’s list they wish (number their choson people from top to bottom). some countries in Scandinavia use this system.

    Another solution is open primaries, where the party’s list for the election is voted on by rhe people (before rhe general election), NOT chosen or appointed by a leadership/only certain people in the party.

    • My reply was addressing Paul Martin’s third point, in the reply below me, regarding who gets to choose the party’s list under Proportional Representation (should have made that more clear)

Leave a comment...

This site uses Akismet to reduce spam. Learn how your comment data is processed.