Elections are a fundamental component of representative democracy – without them, a country cannot be democratic. For many countries, elections are the main way citizens can influence the programs and policies that affect them.
There are a number of noble goals that societies may want to achieve when implementing an electoral system, but they can frequently be at odds with one another. In this blog, I focus on three: 1) the connection between voter preferences and the final allocation of seats 2) the effects of the number and types of parties that can emerge from these systems 3) how well voters can reward or punish MPs and parties for their performance.
As I have discussed in previous blogs, the UK’s current first past the post (FPTP) system and proportional representation (PR) are frequently on the opposite ends of a spectrum. FPTP systems can result in a significant disconnect between the votes a party receives and the final makeup of the legislature as a whole. The resulting parliaments have relatively few political parties in them, calling into question how accurately the legislature truly represents voter preferences. Voters may also be incentivised to vote tactically by casting their ballot for a candidate that has a reasonable chance of winning rather than the candidate they truly prefer.
The fewer and larger parties that result from FPTP mean that governments tend to be easy to form, though. Indeed, single party majority governments are the norm in Westminster. These governments are generally stable and often able to create and implement legislation comparatively quickly. These types of governments also make it easy for voters to know who to reward or punish when the next election comes along because it is clear who is responsible for government policies.
In contrast, PR systems, particularly those with high district magnitudes (the number of MPs elected per constituency) produce legislatures with a closer alignment between a party’s vote shares and the final seat shares. However, the proliferation of parties typically elected in these systems can make forming a government extremely challenging, if not impossible. If a government does form, it’s typically composed of numerous parties, meaning that policy making can be cumbersome. These governments are also generally less stable because should any one party withdraw, the government as a whole falls. This can result in a cycle of frequent elections, with each one followed by extended periods of gridlock and declining state capacity. Finally, multiparty governments make it difficult for voters to identify who is at fault for a government’s failures or successes, making it harder to know which party they should reward or punish in the next election.
Faced with the tradeoffs associated with FPTP and high district magnitude PR systems, some countries have tried implementing other electoral systems to achieve some form of middle ground. However, in order to gain the benefits of one system, you must almost always lose the benefits of another.
PR with lower district magnitudes or thresholds
One of the simplest ways of reducing the chances of a highly fragmented legislature under PR is to implement measures that prevent the very smallest parties from gaining seats. Lowering the “district magnitude” so that fewer MPs are elected per constituency introduces a level of disproportionality, favouring larger parties. Countries can also introduce a minimum vote threshold, dictating a minimum percent of votes a party must receive in order to gain any seats at all in the legislature. Both of these methods reduce the likelihood of a highly fragmented party system, but at the same time create more of a disconnect between voter preferences and the final makeup of the legislature. Additionally, supporters of the smallest parties may feel a need to engage in tactical voting when their parties are in danger of not meeting the threshold.
Majority runoff
Under first past the post, a candidate only needs to receive more votes than anyone else to win. If there are more than two candidates, this means that a candidate can be elected even though a majority of people (more than 50%) voted for someone else – they just didn’t all vote for the same someone else. Majority runoff systems attempt to remedy this concern by ensuring the final winner has the support of a majority of voters. Voting takes place in up to two rounds. A candidate can win in the first round if they receive 50% or more of votes. If no candidate has a majority in the first round, the two candidates that receive the most votes compete in a second round of elections.[1] Mathematically, this ensures that the final winner will have more than 50% of votes, unless the two candidates are absolutely tied. Across the world, majority runoff is the most common presidential electoral system with countries including France, Poland, Chile, Brazil, and Argentina employing it.
Critics of the majority runoff system argue that the majorities winning candidates receive are frequently manufactured rather than accurately representing voter preferences. Furthermore, incentives remain for tactical voting during the first round because voters will only want to support candidates that have a chance of entering and then winning in the second round.
Like first past the post, the fact that only one candidate is elected per constituency gives an advantage to larger parties, though it may allow for more parties than under a pure FPTP system because there is more opportunity for strategic cooperation between smaller parties. Governments formed under majority runoff systems tend to have only a few parties at most, allowing voters to easily identify which parties to punish or reward.
Alternative vote (AV) aka instant runoff
The AV system aims to address the same issue with FPTP as majority runoff – ensuring that the winner has majority support – without having to hold two rounds of elections. Rather than casting a vote for a single candidate, voters fill out their ballot by ranking candidates by order of preference. By doing so, voters essentially indicate who they would vote for in a second round if their most preferred candidate(s) had not advanced. A candidate wins if they receive more than 50% of first-preference votes; if no candidate has a majority, the candidate with the fewest first-preference votes is eliminated. The ballots ranking the eliminated candidates first are then redistributed to whichever candidate the voter ranked second. This process continues until a candidate with a majority appears.
While tactical voting is technically possible under the AV system, it is very challenging to do so effectively. Nonetheless, a single district magnitude still favours larger parties because too few voters rank small parties highly enough. Like majority runoff, AV systems are subject to the same criticism regarding manufactured majority support for the person ultimately elected. In addition, accountability for individual MPs and parties can become more complicated under the ranking system than when purely voting for an explicit candidate.
Under FPTP, the candidate that the most people prefer wins. Under AV however, the winning candidate is frequently the one that the fewest people dislike. In other words, a compromise candidate. Of course, the UK’s own referendum on introducing AV at Westminster elections failed in 2011, with only 32.1% of voters supporting the change. There, opponents pointed to a greater chance of coalition governments in the future and the increased probabilities of extremist parties such as the BNP gaining seats in parliament. It is worth noting, however, that we have not observed these potential pitfalls in a significant sense in Australia, the main parliamentary system that uses AV.
Single transferable vote (STV)
STV systems aim to address concerns regarding the lack of proportionality under AV while still allowing voters to express more nuanced opinions than they would be able to under most PR systems. Like AV, voters complete their ballots by ranking candidates rather than choosing a single candidate or party. Unlike AV, however, more than one candidate is elected per district. Instead of needing majority support, candidates must meet a quota of votes, based on the district magnitude. If there are still seats available after the first round of counting, vote redistribution comes both from removing the candidate with the fewest votes but also proportionally redistributing votes for candidates who have already met the quota based on how many “extra” votes beyond the quota they received.
Like AV, STV systems often advantage compromise candidates. Indeed, rather than gaining their seats through first preferences, the neutral Alliance and Green parties in Northern Ireland often win seats in the Assembly because both Republican and Unionist voters will rank them in the middle. For example, in the 2022 election, Sinn Féin’s first preference shares roughly matched its final seat share in the Assembly (29% of votes and 30% of seats) while the Alliance Party received 18.9% of seats with only 13.5% of first preference votes.
Because more than one candidate is elected per district, smaller parties have more of a chance of entering the legislature in an STV system than they would under systems with a district magnitude of one. However, the need to rank candidates means that STV is only viable with relatively small district magnitudes – it simply would not be reasonable to ask voters to rank many dozens of candidates in an informed way. The small district magnitude (averaging around four in Ireland, for example) introduces disproportionality between voter preferences and final seat allocations but also means that governments will tend to only require a few parties to achieve majority support.
Mixed member/additional member systems
Despite the disadvantages of single district magnitude systems, there can be advantages when each voter has a single MP they know is theirs. Most notably, it is clear who voters should go to for constituency-related issues. If there are multiple MPs per district, each one can take a “not my problem” approach and shift responsibility or blame to another MP representing the same constituency. Voters may also feel less of a connection with their MPs when these MPs come from multiple parties all supposedly representing a single set of voters.
Mixed member systems attempt to keep the advantages of having a single constituency level MP but still allow the greater representation of parties that PR systems allow. Voters in mixed member systems must vote in two different ways: first, by choosing a single candidate that will be elected via FPTP and second by selecting a party that will be allocated seats proportionally.
Any candidate that wins the most votes in the FPTP tier automatically gets a seat. Seats can be distributed from the proportional tier in two different ways, depending on whether the tiers are connected. Mixed Member Proportional (MMP) systems, used for both the Scottish and Welsh devolved parliament elections, take into account the number of seats each party wins in the FPTP tier and awards PR seats to ensure the final seat shares for each party in the legislature as a whole matches the proportion of votes they received in the PR tier. In contrast Mixed Member Majoritarian (MMM) systems treat the two tiers independently, meaning that the seats parties gain in the FPTP tier are not taken into account when deciding how to allocate seats from the PR tier. As its name suggests, MMP systems allow for greater proportionality and therefore more parties in the legislature, for all the goods and ills that brings.
Conclusion
The very fact that so many democratic countries use different electoral systems indicates the challenges associated with the goals and consequences of these systems. Of course, the choice of electoral system is not independent from those who are elected by them. Indeed, officials elected under a given system are frequently the only ones empowered to enact electoral reforms. But why would an elected official want to change the system they were able to win under? As a result, electoral systems are difficult to change, even when it becomes clear that a society’s overall preferences or goals would be better served by doing so.
Kathryn Wainfan.
Kathryn Wainfan holds a PhD in Political Science from the University of California, Los Angeles. She has been a lecturer in Political Science at the University of California, Los Angeles, University of California, Merced, and Loyola Marymount University.
The Constitution Society is committed to the promotion of informed debate and is politically impartial. Any views expressed in this article are the personal views of the author and not those of The Constitution Society.
[1] The version of majority runoff I address here is sometimes referred to as a “majority-majority” system. Another variation, known as “majority-plurality” exists in which the second round does not only include two candidates but instead all candidates that reach a certain threshold in the first round. Thus, only a plurality of votes is required in the second round, not a true majority.