A Return to Normality? How the Electoral System Operated

At first glance, the 2015 General Election heralded a return to normality. According to its advocates, the single-member plurality electoral system enables the electorate to choose directly between alternative governments by ensuring that whichever party comes first in votes secures an overall majority in seats, even though it may have won much less than half the vote.1 For most of the post-war period that is precisely how it has operated. Now, after a hiatus in 2010, when no one party won an overall majority and the partisan colour of the government was determined by post-election coalition negotiations,2 one party, the Conservatives, was returned with an overall majority despite winning just 37% of the UK-wide vote. Indeed, the party secured a majority even though, at 6.6 percentage points, its lead in votes over the Labour Party (in Great Britain) was 0.6 of a point less than it had been five years previously. The system can apparently be relied upon after all to reward the winning party with enough of a ‘bonus’ in seats to ensure that it wins an overall Commons majority. However, a closer look at the result suggests in many respects the electoral system did not deliver what its advocates often claim as its merits. For a start, although the Conservatives won an overall majority, it was by historical standards a small one—just 12 seats—and certainly not one that can be guaranteed to withstand the potentially chilly winds of by-election losses and defections to which all governments tend to be subject. The party's majority is small even though the 6.6-point lead enjoyed by the Conservatives is larger than that secured by Tony Blair in 2005 and Edward Heath in 1970. Yet those two administrations enjoyed overall majorities of 66 and 31 respectively. The Tories' lead in votes is larger too than those enjoyed by the Conservatives in 1955 and 1959, yet those administrations also had more substantial parliamentary majorities of 59 and 99. Meanwhile, apart from delivering a winner's bonus, another key feature of the way in which single-member plurality is meant to ensure elections are a choice between two alternative governments is by severely limiting the representation of third parties in the House of Commons.3 Yet, despite a collapse in support for what had hitherto been Britain's principal third party, the Liberal Democrats, for the third time in a row more than 80 MPs were elected from parties other than Conservative or Labour. The new Parliament will be as fractured as its immediate predecessors. Evidently the way in which the single-member plurality system worked in 2015 merits closer examination. This chapter assesses, first of all, how the system treated the two largest parties, the Conservatives and Labour, and explains why the majority it gave the Conservatives was so small. It then examines the way in which the system treated the smaller parties. Finally, the chapter considers the implications of what happened in 2015—and what might have happened if an alternative, more proportional system had been in place—for the debate about electoral reform. The degree to which the single-member plurality system rewards the winner of an election with a bonus in seats depends on two analytically separable aspects of the way in which it rewards the votes cast for the two largest parties.4 The first is the degree to which the system exaggerates the lead in votes of the largest party over its principal rival, but does so in a manner that is independent of whichever party is the winner. The second is whether it treats one of the two largest parties more favourably than the other. The system may fail to deliver the winner a large majority either because it does not exaggerate the lead of any party to any significant degree, or because it treats the winning party less favourably than its principal competitor. The presence or absence of both features depends on the way in which support for the two largest parties is distributed across constituencies. The system will only provide the winner with a substantial bonus if a relatively large number of seats are highly competitive (or ‘marginal’) between the two parties.5 In those circumstances, seats readily change hands from one party to another, thereby making it likely that even a party with quite a small lead in votes will enjoy a substantial lead in seats. If, however, there are relatively few such seats, then a party might need a big lead in votes before it secures a majority of seats. Meanwhile, the system will only treat the two largest parties equally if two conditions hold (or at least counterbalance each other).6 First, the seats won by one party should on average contain just as many voters as those won by its principal opponents. If that condition does not hold, and a party is more successful in constituencies that have fewer voters, that party is likely to win more seats for any given share of the overall national vote. Second, one party's vote should not be geographically more efficiently distributed than that of its opponent. A party's vote is efficiently distributed if it wins a relatively large number of seats with small majorities and, conversely, loses relatively few seats narrowly. Efficiency also includes not registering high levels of votes in constituencies that are won by a third party. Table 2.1 presents a variety of statistics that help us assess the ability of the single-member plurality system to exaggerate the voting lead of the largest party over the second party irrespective of which party has come first. On the right-hand side, the table reports two statistics that summarise how the division of the votes cast for Conservative and Labour alone (or the two-party vote) is distributed across constituencies. The first is the standard deviation; the bigger the number the more the two-party vote varies from one constituency to another. The second is the kurtosis, which provides an indication of whether there is a relatively large or small number of constituencies in which the two-party vote is relatively evenly divided. The more negative the number the fewer such constituencies there are. Changing distribution of the two-party vote, 1955–2015 Source: Curtice, J., ‘So What Went Wrong with the Electoral System? The 2010 Election Result and the Debate about Electoral Reform’, Parliamentary Affairs, 63, 623–638, and author’s calculations. Marginal seat: Seat where Conservative share of two-party vote (overall Conservative share of two-party vote −50%) lies within the range 45−55%. cast for Conservative and Labour Table only on seats won by Conservative or Labour at that election and by both parties. Changing distribution of the two-party vote, 1955–2015 Source: Curtice, J., ‘So What Went Wrong with the Electoral System? The 2010 Election Result and the Debate about Electoral Reform’, Parliamentary Affairs, 63, 623–638, and author’s calculations. Marginal seat: Seat where Conservative share of two-party vote (overall Conservative share of two-party vote −50%) lies within the range 45−55%. cast for Conservative and Labour Table only on seats won by Conservative or Labour at that election and by both parties. On the side, two statistics the implications of two statistics are the number and the of seats that can be between Labour and the Marginal seats are as those in which the Conservative share of the two-party vote be between and as a result of a of votes across all the two parties were to enjoy the share of the vote across the as a are in other the seats that be most competitive between the two largest parties in the of a overall the fewer such seats there the less of a winner's bonus the electoral system is likely to The division of the two-party vote has to a from one constituency to another than it did in the and The has also been more negative than it was those a there have been fewer seats. This was the result of a Labour in the and more half of the and a to the Conservatives in the and more a that was only in the of electoral However, it that the of seats has more and is as as it has seats within than though as a of all those seats won by either the Conservatives or Labour, the is much the as it was in Meanwhile, although the standard of the two-party vote is from 2010, the is at an This in the number of seats the that a less than on the merits of the two largest parties. and at Conservative support in constituencies with relatively levels of because were that are more likely to have the of the In any such are where the Conservatives were relatively in the first a the Conservative share of the two-party vote on by just of a point in seats in and that the party won in 2010, but by in seats that Labour won time If the to in two then there will be fewer where the is between the principal in the number of seats it more than less that the Conservatives should have to win an overall majority, small. why were to need to look at the second of the way in which the electoral system treated the two largest parties, that is whether it treated one party more favourably than another. help us what might have happened on that Table reports two more statistics that summarise how the division of the two-party across The first is the between the average share of the two-party vote won by the Conservatives across all seats in Great and the overall share of the two-party vote won by the Conservatives across as a two will if one of the parties to in constituencies where fewer In if there tend to be more voters in constituencies where the Conservatives relatively the share of the vote won by the Conservatives will be than the party's overall of two-party 1955–2015 Source: calculations. on of what the have been if that election had been on the new constituency that were at the The with a in the number of was to cast for Conservative and Labour on all seats in Great of two-party 1955–2015 Source: calculations. on of what the have been if that election had been on the new constituency that were at the The with a in the number of was to cast for Conservative and Labour on all seats in Great The second is the between the share of the two-party vote won by the Conservatives and the vote the party won across all seats. The is the share of the two-party vote that constituencies two with one half those constituencies where the Conservative Party won more than the vote and in the other where it won If a party's vote is efficiently its vote will be than its if a party wins seats with relatively small will tend to its vote making or no to its of Table that the first of has to the Labour For much of the post-war Britain's has to of Labour the half of the Conservative and (in the the constituency in to that are Meanwhile, and both where the Conservatives less also tend to be to share of the the the Conservative share of the two-party vote has to be the overall but that after the has each time the constituency have been The Conservatives had that the parliamentary constituency be before the 2015 and that time the be more to as as be to that the of and In however, were when coalition the Liberal Democrats, with Labour to the by that time was in the by of coalition for of the House of In the of Britain's has less in the between the average electorate in seats won by Labour (in and that in constituencies by the Conservatives did from at the time of the 2010 election to However, between constituencies in the number of voters are not the only that can about on of in the of the can also because of in the of at other the in constituencies won by Labour was than in those by the However, the was a less than it had been five years previously. This at the average in seats won by the Conservatives was than the for constituencies won by Labour The in 2010 was The of two was in more votes were cast in the average constituency won by the Conservatives than in that won by Labour, the as in if the in the average number of votes cast in Conservative and Labour constituencies was much the as it had been in 2010, the was not when it to the of the distribution of the two the of had the Conservatives, not least because Labour majorities in constituencies with of not least it has been vote that has been the more efficiently Indeed, it was as a result of with the that the party to from in the of that that at elections the system treated Labour much more favourably than the Conservatives by the third of Table which the two of by the and the overall Conservative share of the two-party vote) Labour in to win a overall majority of 66 in even though the party only enjoyed a lead in of Now, however, has apparently been The Conservative share of the two-party vote in 2015 was no less than 2.1 than the a as the third of Table is more than enough to counterbalance the of the between the of Conservative and Labour constituencies. is one key for The Conservatives in those seats that won most in 2010, and which the party might have been to given that across and as a there was overall a from Conservative to Labour. In in those seats that the Conservatives won most over Labour in 2010, there was a to the in seats that the party won with less than a majority, was of the of in the to that least of the for is in with the at Conservative MPs were elected for the first time in 2010 after an Labour and had had the in the five years to party. On such new Conservative MPs enjoyed no less than a in a the Conservatives a of just two seats to Labour of the of seats that the party have if there had been a to Labour in in and which the Conservatives not have won an overall majority. Meanwhile, the that Conservative in seats had on the of the distribution of its vote is by the that of the seats that as no fewer than be won by the Conservatives in the of an even division of the vote, just be won by Labour. In the Conservatives are winning more seats with relatively small However, there is one change in the distribution of Conservative and Labour support of which the statistics in Table not This is the number of votes that each party in seats that are won by parties other than Conservative or Labour. at many more of the seats that the Liberal have won have been in which principal were the Conservatives than Labour, has hitherto been a of for the at the collapse in Liberal support the Conservatives to no fewer than seats from the Liberal Democrats, more than as many as the 12 seats that Labour from that that were to the ability to win an overall majority despite a smaller lead in votes over Labour than in are just seats in which the Conservatives are second to a Liberal At the Labour seats to the in fewer than of the seats that the party was to in were by the In two that it was Labour that was more votes in constituencies that were won by third parties. On Labour won of the vote in seats won by third parties, the Conservatives secured just a of the in 2010 when the Conservatives on average won in such seats, Labour just The Conservatives were to win a small overall majority for two First, the party from the in Liberal so that votes that had been second to the Liberal were to Conservative Second, an average in seats, not least to the of Conservative MPs seats for the first that the party's vote was more efficiently distributed and fewer seats to Labour than have been the In the party won a majority not because the electoral system to exaggerate the lead of the largest party over the second party, but because of the of of the of that the party had hitherto support for parties other than Conservative or Labour from (in Great Britain) to but is than at any election between and 2005 the parties in won seats in the House of a second only to the seats won by such parties in and on the such seats in However, hitherto it has been the Liberal that have such as the third party in the House of Commons was by the how the of the system to third parties In just as the ability of the single-member plurality system to provide the winner in votes with a bonus in seats is on the distribution of party so also is the to which the system to reward smaller If a party's vote is relatively evenly the system will reward a smaller party with few if any seats, as it it will a relatively small share of the vote On the other if a smaller party's vote is in then it may be quite by the The and potentially way in which the system treats smaller parties was by the 2015 The system did more or less parties. which third in votes across the as a with the parties as as the and the Liberal in of seats. It was the party that in the that in the third largest party in the House of The between the two parties was that vote was for the most relatively evenly support for the of to constituencies in in winning of the vote in Great was It was the first time the of the in that the Liberal before the Liberal had been in a General Election from as the third most party in of It as the most significant independent in the party was with just one the of the from the Conservative At the standard of its support across constituencies was much less than that of the Liberal The party only to win more than a of the vote in and a second within of the in just the vote is to and within in within that of the its vote is geographically quite evenly In 2010, for the standard of support was just in that for the Labour Party in was no less than the for the Liberal of the at also much winning a bigger share of the vote than the Liberal Democrats, the party won just seats with the Liberal of an evenly vote is for a party when it is relatively it an when it is is the in which the in 2015 in winning half of the vote in Indeed, the vote to be even more evenly time with the standard to This change was in at least the result of a if given the of losses of the vote by across as a in Labour did not win as much as of the vote in This meant that to at least vote more in seats where the party was relatively often the relatively This is precisely what Table for the most vote more the vote more Labour had been in The only to was in seats where Labour won between and of the vote in 2010, are the of where Labour to its share of the vote, in all to a high of to Labour by Conservative and Liberal in Labour and support between 2010 and 2015 in by the of Labour support 2010 Source: calculations. in Labour and support between 2010 and 2015 in by the of Labour support 2010 Source: calculations. a result of vote in much more evenly the standard of its support to just that it was by a way only the second most party in to ensure that the party's losses and the were as high as if the in Labour support and the in the vote had been across the of Labour have as many as of its seats to the but in the its of losses to be as many as The of what to the distribution of a party's vote when it loses are also in the of the Liberal The party's vote as across as a did within there were constituencies where the party did not to win as much as of the vote in In it was that to degree at least support for the party should more in where it had been thereby the to which its vote from one constituency to another. is what In those seats where the party won less than of the vote in 2010, the party's support on average by On the other it by in those where it secured between and in those constituencies where it had with more than of the vote its vote on average by as much as of the features of the party's at elections had been a of in of thereby ensuring a more geographically vote that more when it to winning of that was At not only was the standard of party support on the of in 2010, but in it was than at any election that the party all of its of It has often been that Liberal MPs are on in to win and constituencies. Indeed, the party that on of its MPs to seats the national were not in that In those seats where the Liberal was to or the party's vote on average by just points, the point that the party in seats where it was relatively all that meant was that the vote for the Liberal by more or less as much as the party's support did across the as a Indeed, the party's of seats, its was two seats fewer than it have been if the change in support for the party had been across the its average share of the vote in the seats it in Great Britain) from in 2010 to a in the no more than the that first won in The party's vote evenly with a standard of just from the party to win over of the vote in just one other The party was way apart from the and of support to a substantial parliamentary its vote was much less than that for in to the seats it or that for the principal parties that seats in all of won more seats than either the or a result of the collapse in the Liberal vote and the in the way the electoral system treated the smaller parties, the 2015 election has a change in the of representation in the House of it was by a party, the Liberal Democrats, that had a substantial vote, one that had more geographically than previously. Now, representation is the of parties that only elections in or and which are to the and of those of the is the of an electoral system that is to the as as the of a party's of the single-member plurality system often on the of the that the system can Indeed, on one the of the between the of the vote and of the seats by at the of the 2015 election was as as that of any post-war the in was more if of proportional representation had been in such as the party system in elections to the the Conservatives not have had an overall majority, though might have have been the a of a on its of the if had been to come to an on a government with of the 2015 election proportional representation Source: calculations. by number of seats in each proportional to of seats within each determined by but to those parties in a that won at least of the vote. of the 2015 election proportional representation Source: calculations. by number of seats in each proportional to of seats within each determined by but to those parties in a that won at least of the vote. However, of the electoral system the It is not meant to that are as it is meant to be such that the winner secures an overall majority and smaller parties despite the of the Conservatives in winning a small overall majority, it is whether the system can be relied upon to those that the the Conservatives were to win an overall majority not because the system more at a but because of the in which the party was treated less favourably than Labour were no Indeed, there are fewer seats than Meanwhile, the number of MPs is more or less as high as it has This does not a for ensuring that elections will an overall certainly not of the that were for most of the post-war is the when how many seats each party might win Table presents what the be if the and distribution of support for third parties as it was in but the Conservative lead over Labour varies as a result of of support from the in 2015 that in each and for the first row of the table what as a result of a of support from Conservative to Labour that in each and the Conservative lead from the 6.6-point lead the party enjoyed in 2015 to one of The the of larger that is the that result in Conservative and Labour the share of the overall vote, the just over that Labour of the Conservatives, and the that the party was The between seats and votes the 2015 General Election Source: calculations. The between seats and votes the 2015 General Election Source: calculations. key from of all, the system does treat the Conservatives more favourably than it does Labour. On Labour have fewer seats than the Conservatives if the two parties were to win the share of the vote and need to be before it won the number of seats. Labour need to be as much as of the Conservatives to win a overall majority, the Conservatives need a lead of only in to the least of the for are the losses that Labour in on the in the result in Labour just one from the Second, the range of that result in a Parliament in which no one party have an overall majority is between a Labour lead of and a Conservative one of result in no party winning an overall majority. This range is than it has been after any In it that there are about the ability of the electoral system to an overall majority in the that either the Conservatives or Labour only enjoy a In the election will not be the parliamentary that have a parliamentary majority it can be that the Conservatives will in a of the to a successful the and the Conservative that will also be by a in the number of MPs from to On its the is likely to ensure that the electoral system treats Labour even less favourably that it does at by the between the average number of voters in Conservative and Labour constituencies. though will not ensure that the system is in its of the two largest parties. However, the of the Commons to can be to to a degree at least the to which the system a larger constituencies are more likely to be in and and more likely to be competitive between Conservative and Labour. It however, that change will be to ensure that the electoral system the ability to an overall majority in all but the of circumstances, not least because it will to the substantial of The may have another Parliament time but it at of one in even if single-member plurality in