It could be argued that SMP is the simplest electoral system to have ever been conceived. An understanding of how it works can be attained simply from its name. The fact that it is a ‘Single Member’ system demonstrates that voters are choosing only one candidate. SMP is, therefore, described as having a District Magnitude (DM) of 1. The word ‘Plurality’ suggests that the winning candidate does not need a majority of the votes in order to secure a seat but rather a plurality; just one more than their nearest rival. This sets SMP apart from most other non-proportional systems[1] which generally require an actual majority, or 50% of the vote plus 1. Farrell (2001) explains that “the plurality system is said to exaggerate the winning party’s lead, making it easier to win a clear majority of the seats” (p.29) which in turn should, according to the philosophy behind SMP, establish governmental stability. The fact that disproportionate majorities are formed through SMP can be clearly demonstrated in Table 1 which shows the difference between percentage of seats won relative to votes won in UK elections from 1945-1997.
Table 1: Votes vs. Seats in UK Elections from 1945-1997
(Reference: Farrell 2001 pg.27 highlighting added) Note: Statistics highlighted in yellow are significant at a later stage of this paper.
This simplicity of voting, counting and assigning of votes makes SMP incomparably simple to use and makes other systems, such as STV, seem overly complicated. Tideman (1995) explains that the STV concept of voting was first devised, if only in embryonic form, in 1819, by Thomas Hill “for electing a Committee of the Birmingham (England) Society for Literary and Scientific Improvement” (p.29) It gradually took shape, adopting a variety of, even now, flexible rules. Some of the key players in its development were people such as Hare and Andre (Farrell, 2001). It was a century after its first conception that the STV was introduced into Irish politics and was properly implemented for an election a year later in 1920 (Ogg 1920). Tideman (1995) also explains that “STV is the predominant form of proportional representation in countries with a heritage of English influence” (p.27)
Although slightly more complicated than the very simple SMP, STV is a fairly straight-forward system. The electorate are presented with a ballot paper offering them a choice of candidates. The main distinction between STV and SMP is that more than one of the candidates listed on an STV ballot will be elected as a representative. Farrell (2001) describes this ballot structure as ‘ordinal’ whereas the ballot structure for the SMP system is ‘categorical’ where the choice is ‘either or’. The UK uses SMP, as described above, and, therefore, has a DM of 1, meaning that each constituency has only one MP whereas in Ireland the DM ranges from 3-5. The voters use numbers to rank the candidates in order of preference placing a ‘1’ next to their favourite and so on. Farrell (2001) points out that, although in Ireland a voter can vote for as few or as many of the listed candidates as they like, other countries that use the STV system, such as Australia, sometimes require voters to rank every single candidate in preferential order[2]. The STV process, so far, has been pretty straight-forward for the voter. It can prove a little more difficult when one is presented with a list dozens of candidates but in Ireland the list is usually fairly short, especially as their highest DM is only 5.
Then begins counting, a process which, as Farrell (2001) acknowledges, can take days. The first round is simple. The first choice candidate of each voter is identified, giving each candidate a value of total votes accrued. Although it is then possible to list all the candidates in an order of preference the process is not over. Only candidates who have gained a certain quota of votes automatically gain one of the seats on offer. The quota most often used is called the Droop quota. Farrell (2001, p.127) illustrates the formula used in the following way:
To give a simple example, if there were 1000 valid votes for 4 contested seats one would have to divide 1000 by (4+1) or 5 and then to that add 1 more single vote to the sum. That would mean a seat, in this example, is won when a candidate secures 201 votes. If, as is often the case, not all the seats are filled by candidates who have exceeded the quota, the counting must enter a second round. The first step of the second round is to redistribute the surplus votes of any candidates who have already won a seat. These votes will be redistributed to the second choice candidate of all the people who voted for the already victorious candidate. Farrell (2001) explains that, as it would be unfair to pick these surplus votes randomly, a formula has been devised to redistribute votes more fairly. The number of surplus votes is divided by the total number of votes won by the already victorious candidate. This number is then multiplied by the number of second choice votes received for each remaining candidate, giving each an allocation of votes. If all seats are still not filled and redistribution of a second winning count is unlikely to fill all of the available seats then the candidate with the least votes is removed from the contest and their votes are redistributed in a similar fashion as explained above. This process is repeated until all seats are filled.
It is clear from these descriptions that SMP is the most straight-forward and easy-to-use system. What must now be discovered is how significant the ease of use is when compared with the other desired outcomes of electoral systems. A survey of electoral systems experts[3] (Bowler et. al. 2005) was conducted to establish “expert opinion” on a variety of issues, including the most desirable outcomes for an election. The potential outcomes of the different electoral systems tested in the survey were: representation of minorities, accountability of government, proportionality, effective government, good constituency representation and cohesive parties. These properties were then voted upon by the specialist academics to determine which outcomes were most desirable and which were least important. Table 2 shows the results of this survey.
Table 2:
Reference: Bowler et al. (2005) p.11
Although STV fulfils the most highly ranked criteria, namely proportionality of outcome, it could be accused of not fulfilling the fourth most popular category; simple for voters to use. It is democratically important that an electoral system is easy to use since, if it is not, it may be said that only the well educated will be able to fully understand the system and, therefore, express their vote accurately. This could quite easily lead to a disproportionate over-representation of the educated classes in Parliament. In praise of the simplicity of SMP, Farrell (2001) states: “The whole process requires barely a minute to complete” (p.23). May 2007 witnessed much controversy surrounding the use of the STV in Scottish elections. The BBC (2007) and many other news agencies reported the occurrence of spoilt ballots because people did not fully understand the system. It could be said, however, that the confusion did not exist because of the electoral system but rather because people were asked to perform two different elections with two different systems at the same time. In defence of STV, it could be said that, although the system of counting and distributing votes may be fairly complex, the role played by the voters themselves is very straight-forward. In fact, because each vote under STV can be so influential, the effect each voter’s choice has on the overall political situation may be more easily understood than in SMP where tactical voting must often be used to make any political difference. Referring to elections under STV in Ireland, Farrell (2001) quotes Bogdanor as saying, “the system [did not] prove complicated for voters” (p.29). He went on to show that there was not an increase of invalid votes in Ireland when they switched to STV.
It could be said that the two most popular outcomes, as shown on Table 2, are also the two most likely not to co-exist strongly within a single electoral system. It is generally boasted that non-proportional systems, such as SMP produce stable governments, while proportional systems, such as STV, clearly produce a proportionality of outcome and, arguably, a less stable government. Bowler et al. (2005) explain that the reason these two, seemingly, polar opposite outcomes rank highly is that “people who like [proportional representation systems]… do not like SMP [and other non-proportional systems” (p.11) and as academics with a variety of opinions were polled, a division in opinion over outcome emerged. Those academics who favour stable government as an outcome will support systems such as SMP, while people who desire proportionality as an outcome will favour systems such as STV. While, therefore, it is clear from Table 2 that proportionality of outcome is the most desired feature of electoral systems amongst academics, it is unfair to label SMP as worse than STV simply because it is not proportional. The supporters of SMP never claimed to be proportional and, in fact, do not desire, pretend or want it to be such.
The purpose of SMP is to produce “one-party majority governments that would otherwise not exist” (Blais and Carty 1988, p. 550) It is thought, therefore, that when SMP achieves that it produces stable government which is considered more important than any of the other desired outcomes. If, however, it could be demonstrated that SMP neither produces one-party majorities nor, even when it does, will it ensure a stable government, the supposed strengths of the system seem to vanish. Blais and Carty (1988) go so far as to say that, “without this benefit, the seat bonus that [SMP] regularly delivers to large parties could not be easily justified” (p.550) Blais and Carty (1998) explain the following:
“There is no doubt that [SMP] elections are more likely than proportional ones to assure a majority to a party with only a plurality of the vote. The relative probability is nine times as high. But the data indicates that such outcomes are even more common in multi-member majority contests.” (p.552)
It seems clear from this that SMP is not even the best system among non-proportional systems at producing the results its supporters suggest are most important, namely strong, stable governments. Combined with this are some interesting points we can see from looking back at the highlighted sections of Table 1. It is clear that on at least five occasions, between 1945 and 1997, SMP in the UK has only managed to secure a very weak majority in parliament. It could be said that this produces weak or fragile governments, a direct antithesis of the purpose of SMP. So, not only is SMP statistically not the best system for producing stable government, in practical terms it has failed 5 out of 15 times to produce the strong and stable governments it promises.
Another criticism of SMP is that it is less to do with people and more to do with boundaries. Certain regions of the UK are traditionally supporters of one or another of the two major parties. Individuals living within these constituencies have no opportunity to affect their situation whether they vote or not and, therefore, none of being directly represented in parliament. Closer inspection of the boundaries highlights another issue concerning the population sizes of constituencies and, therefore, individual voters relative values in the electoral system. Farrell (2001) states that, in the UK, under SMP, “the benefit of simplicity can be (and often is) at the cost of fairness – fairness to smaller parties and the supporters of smaller parties, fairness to those ‘trapped’ in seats which are safely held by parties they do not support” (p.29). Delury (1999) gives evidence that “in 1997 [UK constituency populations] ranged from 22,983 registered voters to 101,680” (p.1159) While it could be said that this may be to do with the structure of the country rather than the electoral system[4] it is improbable that disparities as large as this could exist under a system of proportional representation.
It is not only SMP which has technical faults. Tideman (1995) makes various criticisms of STV which seem to apply only when it used to elect a single person. Farrell (2001) would argue that if this were the case the system could not be called STV as “STV operates with a district magnitude greater than 1” (p.126) He does, however go on to highlight some academic criticisms of STV. The main one of these is that STV is ‘non-monotonic’. Farrell (2001) explains that this means, occasionally, someone gaining more votes than anyone else in the first round but not achieving enough to push them past the quota may be damaged by their initial popularity when the votes of excluded candidates are then redistributed amongst the remaining candidates. Farrell (2001) defends STV by highlighting that, “a statistical test of the issue has produced the prediction that, were STV to be used in the UK, there would be ‘less than one incidence [of this] every century’” (p.150).
From the evidence given above we can begin to decide which electoral system is best. The accusations made against STV, that it is too complicated and that it sometimes produces irrational results, all seem to have been discredited or at least identified as relatively insignificant. On the other hand, SMP has been accused of being undemocratic, overly complicated (especially when considering boundary issues in the UK) and it has been shown that SMP does not even do what it is meant to do; produce stable governments. If this is not reason enough one could just ask the experts. The Bowler et al. (2005) survey of electoral system experts asked academics to list electoral systems in order of preference. Overall PR systems ranked 1st, 2nd, 3rd and 5th. Non-PR systems ranked 4th, 6th and 7th. STV came in 2nd and SMP ranked 6th. It is not surprising, given this expert opinion, that when new democracies have emerged in the last century, that they have steered clear of systems such as SMP and rather opted for the more fair and democratic proportional systems, such as STV. Some may be left wondering why the system is not simply changed. The answer seems obvious when we consider the fact that the incumbent party, with whom the power to make changes lies, is unlikely to want to change the system which has secured themselves power.
[1]Although the system is used in Japan, Multi-Member Plurality (MMP) is a very rare form of non-proportional electoral system.
[2] Heitshusen et al. (2005) explain that, when voting for the Australian senate, voters have the choice between regular STV or rather simply voting for their preferred party, referred to as Group Ticket Voting (GTV) which is almost the same as a closed-party list and is far less proportional than STV.
[3] It is interesting to note that throughout the survey the preferred method of vote counting was STV as it seemed to give the most accurate results.
[4] In the USA, for instance, where they also use STV they have much stricter parameters on constituency population sizes which leads to greater proportionality in representation, or gives a more equal spread of representation.
No comments:
Post a Comment