Parliamentary Affairs Advance Access originally published online on November 8, 2005
Parliamentary Affairs 2006 59(1):60-77; doi:10.1093/pa/gsj006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
Second-Class Representatives? Mixed-Member Proportional Representation in Britain
Lecturer in Politics at the School of Politics, Philosophy, International Relations, and the Environment, Keele University
This article compares the constituency roles of the members of the Scottish Parliament and the National Assembly for Wales to their counterparts in two German federal states using postal survey data and interviews. Representatives elected in single-member constituencies spend more time than their list-elected counterparts on constituency service, believing it to be important to their re-election effort, while list-elected representatives focus more strongly on interest groups. Competition between constituency and list representatives in Britain has caused more problems than in Germany. The results suggest that rational-choice considerations motivate the behaviour of politicians, and that the mixed-member electoral system contributes to role differentiation.
DEVOLUTION of power in the United Kingdom to the Scottish Parliament and the National Assembly for Wales has changed the British constitution in many ways. A mixed-member form of proportional representation (PR), known locally as the Additional Member System (AMS), has been used to elect these new bodies in Scotland and Wales. Mixed-member electoral systems are usually characterised by the election of some representatives in single-member constituencies while others are elected from party lists using some form of PR. When the two tiers of seat allocation are linked so that party list seats compensate for the partisan disproportionality arising from constituency seat results, and the overall result on a partisan basis is proportional to the party vote, the system is called mixed-member proportional (MMP) in most scholarly literature.1
It is this mixed electoral method that has aroused more than just the usual British criticism of the coalition or minority governments that are likely to arise after a change to PR. Critics fear that mixed-member PR systems create two ‘classes’ of elected representatives because those indirectly elected from party lists, rather than directly elected in single-member constituencies, will be more oriented towards their parties than towards constituents. The same concern has also arisen in New Zealand where list-elected members, who comprise almost half the membership of the House of Representatives since the introduction of MMP in 1996, ‘have been referred to as "second-class" by the media, parliamentary colleagues and the public alike’.2 In Germany, however, it has generally been assumed that representatives elected from party lists are not tainted with any ‘second-class’ status, although scholarly investigations into the existence of any distinction between the two types of representatives have been infrequent.
Does MMP really create two classes of representatives? To be less normative, does the mixed electoral system reduce the constituency role of those elected from party lists? What kind of constituency role do list-elected members of the Scottish Parliament and the National Assembly for Wales play? This article investigates the role played by representatives elected by MMP in Scotland and Wales, with particular attention to the constituency role. Postal surveys and interviews were used to compare members of these British bodies to their counterparts elected by MMP in two German Landtage, or state parliaments. Representatives are asked about their constituency service activities, which are broadly defined to include contact with interest groups, the seeking of public funds and promotion of the area as a place for business. It is assumed in this study that those representatives elected in single-member constituencies will seek out a stronger constituency role than those elected from party lists.
Questions about the role played by list-elected representatives may appear to reflect a British cultural distaste for multi-member constituencies, but upon further examination, actually articulate a rudimentary rational choice assumption about the behaviour of elected representatives. Rational representatives will anticipate that the constituency service demands of those indirectly elected through list PR will be lower than the demands on those elected in single-member constituencies. This is because list-elected representatives are better able to ‘shirk’ constituency demands, according to the terminology of Shaun Bowler and David Farrell.3 Candidates on party lists are normally selected by regional party organisations and ranked on ‘closed’ lists so that voters cannot influence the ordering done by the parties.
Therefore, those candidates elected through this closed party list mechanism have less of an electoral incentive to serve constituents in their electoral regions because their election depends heavily upon their appeal to the party organisation: the more highly a candidate is ranked on the list, the greater the likelihood of election under PR. For candidates elected from single-member constituencies, re-election depends on re-selection by their party organisation and their appeal to constituents, meaning that they have a stronger electoral incentive to serve constituents. Indeed, a personal vote, cultivated independently of party loyalty, might give candidates in single-member constituencies that extra bit of support they need to retain a seat, making constituency service very important to the re-election process in marginal constituencies.4 Therefore, this ‘electoral incentives’ hypothesis simply states that the way in which representatives are elected affects their constituency roles.
This rational choice approach is the subject of some controversy among scholars because others, like Donald Searing, believe that elected representatives are motivated mainly by their own culturally informed belief that constituency service is simply ‘part of the job’ and valued for its intrinsic merits. One way to examine the relative impact of cultural and institutional factors on the behaviour of politicians is to look at another example of MMP in a different cultural setting. Germany, the model for most mixed-member systems, has used MMP since the late 1940s for its federal and most state elections. If institutional rules are significant in determining how politicians behave, then they should work the same way in different cultural contexts. The widely held perception of political scientists and commentators is that German voters and politicians do not distinguish between constituency and list members; there is no ‘caste’ system in which one type of representative is seen as more legitimate than the other. However, role studies of German legislators that distinguish between constituency and list members have only recently been undertaken.5
 |
The constituency role in Britain and Germany |
|---|
 Top The constituency role in... Testing the electoral incentives...Competition over constituency...ConclusionsNotes |
|---|
Scholars have noted that elected representatives have different roles to play in the political system. Studies by Donald Searing indicate that there are many roles assumed by British members of Parliament (MPs), while Philip Norton and David Wood narrow these roles down to two broad categories or ‘faces’ of MPs: local non-partisan constituency service and national party policy advocacy.6 The introduction of mixed-member electoral systems, requiring the introduction of list PR members, in addition to MPs elected by single-member plurality (first-past-the-post) voting, presents the opportunity for these MP roles to be formally split, allowing a study of how MPs develop their roles along constituency and partisan lines.
British references to party list representatives as ‘second-class’ reflect a received wisdom among many Anglo-Saxon politicians, voters and political scientists that multi-member constituencies dilute the relationship between MPs and their constituents.7 Behind this thinking is the expectation that MPs will act as non-partisan advocates for their constituents, regardless of whether a constituent voted for him or her at the last election. This understanding of non-partisan constituency service was illustrated during a 1999 debate on office allowances by a member of the Scottish Parliament (MSP) who argued that single-member constituency MSPs deserve larger office allowances because ‘problems are not party political’, and voters would bring their problems to their single-member constituency MSP, not to MSPs from the regional list, implying that list MSPs would need less of an allowance. Furthermore, there is a widely held fear among some politicians that representatives elected from party lists will be more accountable to the political party that selects candidates and ranks those lists than to voters. In a 1998 House of Commons debate on the establishment of the Scottish Parliament, an MP argued that ‘First-class [constituency] Members will be accountable for their selection and election to real people, whereas there is some doubt about exactly to whom second-class [list] Members will be accountable’.
The link between British MPs and constituents is seen by most MPs as something akin to sacred, according to some observers. ‘Hard-boiled politicians become almost dewy-eyed about the "sacred trust" and "indissoluble bond" between Members and their local constituents’, writes Ivor Crewe.8 Norton and Wood go so far as to claim that severing the link between MP and constituents ‘would serve to undermine the legitimacy of the House of Commons’. Critics like Crewe, however, argue that the ‘sacred link’ between MPs and constituents in Britain is exaggerated, citing evidence showing that many people do not know who their MP is, and very few can cite something the MP has done for the constituency. Stuart Weir and David Beetham note that only 9.7% of British people surveyed in the late 1980s had any contact with their MP during the previous five years.9 While the Comparative Study of Electoral Systems in the late 1990s found that 13% of British people surveyed reported some form of contact with an MP in the previous twelve months, this level of contact was only about average (12%) when compared to that reported by citizens in eighteen other democracies.10
Despite the arguments of sceptics like Crewe that the ‘sacred link’ between MPs and constituents in Britain is a ‘political myth’,11 however, that myth persists. Even Weir and Beetham admit that opinion surveys strongly suggest that most British people prefer having one MP to several. Therefore, unlike in the past, when the single transferable vote was the choice of PR advocates, most proposals for electoral reform in Britain in recent years have called for the retention of a single-member constituency element. This means that the obvious model is a mixed-member system with party list deputies added to those that a party wins in single-member constituencies to achieve a more proportional result.
In post-Second World War Germany, British occupation authorities believed that they needed to introduce single-member constituencies in some form in order to ‘build and strengthen the ties between the members of parliament and the citizens’.12 Whether Germany’s mixed system of constituency and party list representatives actually does enhance the relationship voters have with those they elect is the subject of some dispute in the literature. Max Kaase, a German critic of the system, argues ‘the initial expectations in designing the personalized PR system with respect to citizen deputy ties have certainly not materialized’.13 In a similar vein, Charlie Jeffery claims that ‘a continuous link between voters and MPs in a particular area does not have the significance attached to constituency representation in the UK’.14
Some authors criticise the electoral system for not being genuinely personalised, while others claim that Germans generally take their grievances to officials other than elected representative. Ultimately, the main observation is that German citizens generally do not distinguish between directly elected and party list representatives, and there is no ‘caste’ system in which one type of representative is seen as more legitimate than the other.15 In fact, dual candidacy (constituency candidates also standing on party lists to ensure election) is common in Germany, with the ‘losing’ candidate in the direct constituency elections often winning a Bundestag seat from that constituency’s Land (federal state) party list. This phenomenon is one reason why Eckhard Jesse states that the voter ‘does not perceive the difference at all’ between the two different types of representatives.16
However, it would be an exaggeration to assert that Germany’s elected representatives are not expected to involve themselves in constituency work at all. German political scientist Werner Patzelt reports that in Germany, ‘it is obvious that for an MP the constituency work is a critical part of holding his office. Consequently, the representatives devote about one-third of their working hours to constituency work’.17 Indeed, Patzelt’s research reveals that constituent demands are so great that Bundestag members ‘are expected to act as ombudsmen with far more rigour than they feel should be expected of them’.18 Furthermore, Burkett points out that German citizens approach both party list and constituency representatives with problems or questions, with both types receiving similar amounts of correspondence. Which representative is contacted by citizens seems to depend on party identification: ‘If "their" party lost in the constituency they may prefer to approach the List deputy of that party, especially if he or she was the unsuccessful candidate of the party in that constituency’, according to Burkett.19
‘Losing’ constituency candidates in Germany who end up being elected via the party list often ‘shadow’ the candidate who defeated them, setting up shop in the same constituency. While shadowing seems to be accepted by German citizens and their representatives (including those who are being shadowed), the practice has brought about problems in Britain. A panel study of thirteen MSPs noted the presence of friction between constituency and list members shortly after the first Scottish Parliament election, often over who is entitled to casework.20 British constituency representatives sometimes complain that the list members will spend time focusing on sexy issues to publicise, cherry-picking the most glamorous cases from their large electoral regions while constituency members are stuck dealing with more mundane local cases.21 Constituency representatives are also unhappy that the shadowing party list representatives are targeting their constituencies for ‘extra’ attention because they plan to stand in that constituency against the incumbent member at the next election.
While most accounts of the German electoral system state that the apparent status differences seen in Britain and New Zealand between party list and constituency representatives do not exist, some scholars now claim that there is at least some difference between the role orientations and behaviour of the two different types of representatives. Recent evidence from Hans-Dieter Klingemann and Bernhard Wessels in the 2000 Shugart and Wattenberg volume shows that party list and constituency members of the Bundestag differ with respect to both the style and the focus of representation, causing the authors to conclude that ‘type of mandate makes a difference for role orientation’. An earlier study by Thomas Lancaster and David Patterson found that ‘district type affects representatives’ perceptions of pork barrel allocations as a special type of constituency service’.22 Patzelt writes that Bundestag members elected in single-member constituencies ‘regard themselves more as successful representatives of the citizens, clearly invest more time in constituency work, and engage much more in case work and constituency service’ than their list-elected colleagues.23 Therefore, the received wisdom in much of the literature that both German constituency and list representatives have such similar constituency roles may be inaccurate.
|
Testing the electoral incentives hypothesis |
|---|
|
TopThe constituency role in...Testing the electoral incentives...Competition over constituency...ConclusionsNotes |
|---|
In order to test whether and how representatives respond to their electoral incentives, a postal survey was used to gather information from members of German and British legislatures below the level of central government. Two German Landtage, those of Brandenburg and Hesse, were chosen because they use the two-vote electoral system used by the Bundestag (along with the Scottish Parliament and the National Assembly for Wales). The survey was a German translation of the questionnaire sent to members of the two British legislatures. German legislators were surveyed in 2000, at the same time as their Scottish and Welsh counterparts, and another survey of the British representatives only was conducted in late 2003. The second British survey allowed for a test of whether representatives ‘learned’ their incentives over time, an interval that included the second (May 2003) election for their legislatures.
While a comparison between two British and two German MMP-elected legislatures seems like a logical way to test the electoral incentives hypothesis, some important differences must be pointed out. Hesse has experienced federalism since 1949, and Brandenburg since 1990, while devolution came to Scotland and Wales only in 1999. The National Assembly for Wales has only secondary legislative powers, while the Scottish Parliament has primary legislative powers over the areas devolved to it by Westminster. The electoral systems, while variants of the MMP model, differ in that proportionality is achieved at the Land (state) level in the German states under study, with list seats allocated on a statewide basis. In Scotland and Wales, proportionality is achieved on a regional basis, with a limited number of list seats allocated on that regional level. Despite these differences, however, the central similarity of members of the legislatures being elected in two different ways remains, and this is what is crucial for the electoral incentives hypothesis.
The overall German survey response rate was 46.6%, with a slightly higher response from Hesse (49.1%) compared to Brandenburg (43.5%). For the British assemblies, the overall response rate was 39.8% in 2000, with a similar response in both Scotland and Wales. The 2003 survey in Scotland and Wales achieved a similar response rate (39.7% overall). The overall response rates are higher than those of other similar investigations, however. For example, Lancaster and Patterson had a response rate of 37.1% from the German Bundestag members they surveyed by mail in a similar project, and William Downs24 had a 35.2% response rate from a postal survey of members of German Landtage (state assemblies). Nevertheless, the fact that these legislatures are relatively small means that there is a small-n problem that must be taken into account when examining the data.
Constituency service was measured both as contact with constituents and as seeking public spending or attracting businesses to the constituency or region. Representatives were asked to indicate the approximate proportion of their work time spent on such activities, within a given set of percentage ranges (0–5, 6–10, 11–15, 16–25, 26–50 and 51–100), and the results were averaged. Because representatives do not punch a time clock when they carry out their duties, there is no way to verify these self-reported estimates, however. The expectation of the electoral incentives hypothesis is that constituency-elected representatives should spend more time helping voters who have problems with government agencies, while their list-elected counterparts should spend less time. This expectation does hold with the German representatives, as summarised in Table 1, part A. While the Germans spend less time on these matters than their British counterparts, constituency representatives spend more of their work time helping constituents with their problems than list-elected representatives (11.4%, as opposed to 9.7%), but the difference is small and statistically insignificant.
|
British constituency representatives surveyed in 2000 also spend slightly more of their work time helping voters (17.5%) than list-elected representatives (15.4%), but results from the 2003 survey reveal a much bigger difference and in the hypothesised direction. In 2003, British constituency representatives spent 21.1% of their work time helping voters, while list-elected representatives spent a mere 13.7% of their work time on constituency service. The large difference, significant at the 0.01 level, between British constituency and list representatives over time spent helping voters in 2003 lends support to the hypothesis that constituency representatives have learned (perhaps from the 2003 election) that constituency service can make a difference to one’s re-election prospects.
While dealing with constituents usually takes the form of holding surgeries (open-house meetings with the public) or answering letters, e-mails or phone calls from regular citizens, representatives may also deal with interest groups. It is possible that list representatives, lacking small geographical constituencies, might seek out a different kind of constituency altogether. In New Zealand, Green party list MP Rod Donald focuses not only upon geographical areas (allocated by the party), but also upon what he calls ‘constituencies of interest’, which include employment, training, trade, tourism and constitutional reform.25 Therefore, while parties may assign geographical constituencies held by constituency representatives from other parties to their list representatives, it appears that interest groups, which represent the constituencies of interest that Donald refers to, might be easier for list representatives to deal with.
The time spent dealing with interest groups is reported in Table 1, part B. Both British and German representatives report spending an average of approximately 13–14% of their work time in contact with interest groups via meetings, letters and phone calls. For German representatives, the difference in time spent on such contacts between constituency and list Landtag members is minor, 14.7 and 13.8%, respectively, but the difference is not in the hypothesised direction. The opposite is true, however, with the British representatives. In 2000, constituency members of the Scottish Parliament and the National Assembly for Wales spent 9.2% of their time dealing with interest groups, while list representatives spent 17.2% of their time on interest group contacts. Almost the same result appears in the 2003 survey, with constituency representatives spending 9.1% of their time on interest group contacts, while their list-elected counterparts spent 16.9% of their time on this activity. Both the 2000 and 2003 results reveal differences between constituency and list representatives that are significant at the 0.01 level.
Another type of constituency service involves the seeking of public spending projects (which may not be efficient ways of spending public money) or the location of businesses in the constituency. This type of activity among German Bundestag members has been studied by Lancaster and Patterson using survey research. Obtaining what Americans call ‘pork barrel’ spending for their constituents can give elected representatives an extra edge over their challengers at the next election, as can bringing new jobs into the area and claiming credit. The 2000 and 2003 surveys asked German and British representatives how much of their work time was spent ‘bringing home the bacon’ and on local promotion, and Table 2 summarises the results.
|
Seeking out pork barrel spending takes up about 7% of the average British or German representative’s time, but this average conceals some major differences between constituency and list representatives. For members of the Scottish Parliament and the Welsh Assembly surveyed in 2000, the differences between constituency and list representatives are slight and are not in the hypothesised direction: constituency representatives spent 6.1% of their time seeking public spending projects for their constituents, while their list counterparts spent 6.6% of their time on this activity (Table 2, part A). In 2003, however, the question, asked again, came up with entirely different results, this time in the expected direction and significant at the 0.05 level. The more recent survey found that British constituency representatives spent 9.8% of their work time seeking pork, while their list counterparts spent only 5.2% of their time on this activity. This result is nearly identical to the finding among German Landtag members, where constituency representatives spend 9.7% of their time seeking pork, while their list counterparts spent only 5.0% of their time this way. This difference is also significant at the 0.05 level.
German representatives also spent an average of 9% of their work time promoting their areas as places to do business, but this average again covers up the fact that constituency representatives spent 12.4% of their work time as local promoters, while their constituency counterparts spent only 6.1% of their time this way, less than half as much (Table 2, part B). The difference is significant at the 0.05 level. The same sort of difference between the two types of representatives is seen among the British in 2003, with constituency representatives spending 9.3% of their time promoting the constituency as a business place, while their list counterparts spent only 4.7% of their time promoting their electoral regions. The difference is significant at the 0.01 level. Again, the difference between the two types of members is in the hypothesised direction, as with the German representatives, but the result from the 2000 survey of Scottish Parliament and Welsh Assembly members was the opposite. British constituency representatives in 2000 spent only 5.3% of their time promoting their constituencies, while list representatives spent 7.6% of their time on this activity.
The German Landtag and British 2003 results in Table 2 are similar to those of Lancaster and Patterson in their Bundestag member research and are consistent with the electoral incentives hypothesis. Perhaps as a result of the intervening election, and simply having more time to ‘learn’ their electoral incentives, British representatives in 2003 responded to the survey in a very similar way to their German counterparts, departing from their 2000 answers on questions about securing public funding for their geographical areas and promoting these places to businesses.
The evidence presented so far suggests that constituency representatives believe that an emphasis on constituency service is important to their re-election effort; otherwise, it seems unlikely that they would spend the time on it. When asked directly whether they believe that helping people who have problems with government agencies is important to their re-election, constituency representatives are, indeed, more likely to say that this effort is ‘very’ important, and the difference between constituency and list representatives is statistically significant in each instance (Table 3). In the British case, there was an increase from 2000 to 2003 in the percentage of both types of representative saying that constituency service was ‘very’ important, but in each survey, the percentage of constituency members is higher.
|
Table 3 shows that this is also the case in Germany, but there is one area that differs from the British results. While 5% of constituency representatives in the German case replied that constituency service was ‘not very’ important to their re-election prospects, a far greater portion of list representatives (22.4%) responded this way. In the two British surveys, only 5.1% of list representatives in 2000, and 2.6% in 2003, said that constituency service was ‘not very’ important, while none of their constituency counterparts responded this way. These results are in the hypothesised direction, but the difference between German and British responses is quite large, suggesting that constituency service has a greater importance in Britain than in Germany, as some of the relevant literature claims.
On the other hand, constituency service is ‘very’ or ‘somewhat’ important to the vast majority of German representatives, both constituency- and list-elected, as Table 3 indicates. Looking into constituency service more closely, another survey question asked whether serving constituents or enacting party policy was more instrumental to their decision to stand for office (Table 4). Here, half of the German constituency representatives said serving the needs of their constituents was more instrumental in their decision to stand, while only 32.7% of list representatives chose this response. Helping to get party policy preferences enacted into law was cited as the motivation to seek office by 55.1% of list representatives. The difference between constituency and list representatives is significant at the 0.05 level. This result suggests that German list representatives prefer an emphasis on policy matters to constituency service, and that perhaps some role differentiation has been brought about by the mixed electoral system.
|
The German pattern of role preference was not found in the 2000 UK survey, in which there was almost no difference between constituency and list representative responses. Over half of both constituency and list representatives in Scotland and Wales claimed they stood for election in order to serve their constituents’ needs, perhaps illustrating the pervasiveness of constituency service in Britain. This is also seen in the fact that only about 40% of German representatives (taking both constituency- and list-elected together) went into politics primarily to serve constituents, while over half of British representatives, both in 2000 and in 2003, said they were primarily motivated to serve constituents (Table 4).
The strong constituency orientation among British representatives appears to be somewhat diluted among list-elected members in the 2003 results, however, with a major shift in the German direction. Among constituency representatives, 62.2% cited serving voters as their motivation for standing, while only 42.1% of list representatives cited this. Helping to enact party policy motivated 52.6% of list representatives, but only 27% of their constituency counterparts. The difference between constituency and list representatives is significant at the 0.01 level (Table 4). This change among list representatives in the hypothesised direction could indicate that more interest in policy than in traditional constituency service follows a learning process—it may have become clear that list representatives, especially those from the smaller parties, will have more policy work to do than their counterparts in the larger parties. Furthermore, the Scottish Greens, who increased their presence in the Scottish Parliament from one to seven members after the 2003 election, do not stand in constituency contests at all.
On the other hand, some list representatives in Scotland and Wales are quite interested in serving constituents because they plan to stand in a single-member constituency at the next election. This is particularly the case for the nationalists (the Scottish National Party in Scotland and Plaid Cymru in Wales), the second largest group in each devolved body. Their activities in constituencies have been very active, as will be described below.
|
Competition over constituency service |
|---|
|
TopThe constituency role in...Testing the electoral incentives...Competition over constituency...ConclusionsNotes |
|---|
Because people in Scotland and Wales now have regional list representatives, in addition to a single-member constituency representative, there is competition over service to constituents. On the one hand, this competition could be good for citizens offering not only more choice, but also perhaps enhancing the quality of constituency service as representatives compete to be the best providers. On the other hand, however, increased competition has not been welcomed by some representatives, especially those who hold the safe constituency seats. In the case of Scotland and Wales, this means Labour MSPs and Welsh Assembly Members (AMs). This section will analyse comments made by British representatives on questionnaires and in interviews.
Competition is likely to be more intense when list-elected representatives target a single-member constituency for ‘extra’ attention in order to get enough publicity to run a stronger campaign in that constituency, if they plan to seek their party’s nomination to stand there at the next election. This ‘shadowing’ is often a source of irritation between constituency and list representatives. When dealing with their list-elected counterparts, constituency-elected members of the Scottish Parliament and Welsh Assembly may claim they are somehow ‘better’ representatives, often invoking the notion that direct election in a relatively small constituency is superior to indirect election from a party list in a large electoral region. This is where the ‘second-class’ label may be used, or implied, by constituency representatives to describe their list-elected counterparts. Even when a class division is not implied, there appears to be confusion about what list representatives do, as this quote from a constituency MSP’s 2003 questionnaire reveals:
List MSPs do not appear to be as accountable to the electorate because they represent such a large area. Some constituents are unaware of the fact that they can approach one of seven MSPs to discuss their concerns. There is an opinion that list MSPs do not have the same constituency demands because constituents do tend to approach their constituency MSP first and foremost.
Other list representatives also pointed out that there is confusion over what constituency- and list-elected members of the Scottish Parliament and Welsh Assembly are meant to do. This comment is from an MSP surveyed in 2000:
I have relatively good relationships with constituency MSPs. I do not try to be a shadow. But I do pick up issues they do not and of course we have significant policy differences. I would much prefer being a constituency MSP for which there is a clearly defined role.
Another MSP in the 2003 survey states simply that members of the public ‘are utterly confused about the role of constituency and list MSPs’.
One potential place for list-elected members to develop a distinctive role is in the representation of regional interests. This should be a by-product of their election on a region-wide basis, as one list MSP suggests in the 2003 survey: ‘List MSPs must develop a broader range of subject areas than single-member MSPs because regions are much more diverse’. Other list MSPs, like this one surveyed in 2003, note this diversity as well:
Because of the diversity of list MSPs’ constituents, civic organisations and lobby groups tend to approach the MSP whom they think will best represent their interests. The management at Prestwick Airport will have links with the Tory MSPs; groups fighting hospital closures will contact the SSP [Scottish Socialist Party]; there is a flexibility.
While list-elected representatives may have stronger connections to interest groups, perhaps seeing them as ‘constituencies of interest’, this kind of relationship may simply not be visible to voters. Indeed, with a much larger area to serve, visibility in the electoral region is more of a challenge to list representatives. This difficulty may lead some list representatives, especially from the larger parties, to shadow constituency representatives they intend to stand against instead.
While many constituency representatives in Scotland and Wales do not have a problem with their list counterparts, others clearly do. Some of the more negative comments, such as from this MSP in the 2000 survey, suggested that there were electoral motivations involved in shadowing:
Unnecessary and against the intention of proportionality. Surrogate constituency MSPs were not the intention of the AMS. Furthermore, the target seats are the ones shadowed. It also suggests a list MSP is not as worthwhile as a constituency one.
Some list representatives, particularly the nationalists, who make up the second largest party in each assembly, aim to take many (mainly Labour) constituency seats at the next election. Because of the multi-member electoral regions (each containing several single-member constituencies within them) created by the mixed-member electoral system, Scottish and Welsh politicians are facing their opponents constantly, even after defeating them in single-member constituency races.
Many list representatives describe what they perceive as ‘second-class’ treatment by their constituency counterparts. Often they confirm that the constituency representatives perceive an electoral threat, as the one list AM, in the 2003 survey, notes:
I do encounter resentment because my work enables me to threaten the constituency AM’s success at the next election.
Sometimes list representatives complain not about individual constituency representatives, but about the party organisation of their counterparts, usually the Labour Party. Here are quotes from a list AM and list MSP, respectively, surveyed in 2003:
No direct friction from AMs elected from my region’s single-member constituencies—but lots from the Labour ‘spin’ machine misrepresenting (lying!) both the legal and constitutional facts and the reality of events in the Assembly.‘Friction’ from MSPs representing single-member constituencies is largely confined to Labour Party HQ. Individual MSPs have no problem with my efforts. Labour with so many constituency reps see list MSPs as a danger.
Labour and the Liberal Democrats dominate the single-member constituencies in Scotland and Wales. Furthermore, Labour governs (with the Liberal Democrats in Scotland), while the nationalists and Conservatives, primarily elected from party lists, are the opposition. This division reinforces the perceived class difference.
Other constituency representatives are less hostile towards the list representatives who shadow them. Instead of resenting it, they see this competition, at least grudgingly, as good for the voters, as this MSP surveyed in 2000 notes:
It can be irritating-to-threatening, but is an inevitable consequence of the system, so has to be accepted. The answer is to work harder and better for the constituency and constituents.
Welsh Assembly constituency AM Kirsty Williams comments in a 2001 interview that because list members target constituencies they plan to contest at the next election, they can be a ‘real pain in the neck’ and even ‘infuriating’, making constituency members feel like they are in a ‘constant state of competition and electioneering’.
This competition between constituency and list representatives benefits voters, who now have more choice, maintains Peter Black, a Liberal Democrat list member of the Welsh Assembly. He cites, in a 2001 interview, cases of people actually approaching their local councillor, MP, constituency AM, and list AMs, and comparing their ‘performance’ with regard to the concerns brought to their attention. Another list-elected representative, this time in Scotland, agrees, saying the following in the 2003 survey:
Electors are now shopping around seeking support from several MSPs. A list MSP has a greater workload but has a better overview of regional issues.
Williams agrees that enhanced choice is good for constituents because ‘you can’t be lazy’, and for the most partisan voters who will never approach a constituency member they voted against, there is the opportunity to deal with ‘their own’ list member:
There may be Conservatives out there who would never come to me in a million years, and why should they be disenfranchised because of a weird electoral system [single-member plurality]? At least now they’ve got someone they feel comfortable going to, and I think that’s fine.
Other representatives also pointed to this advantage of the MMP electoral system in giving representation to voters who have been historically disenfranchised from the standpoint of living in a constituency safe for a party they do not support. One list-elected MSP said the following in the 2003 survey:
I do not frequently encounter friction and I find that there is often scope for cross-party consensus in my region on a number of issues. Obviously this is balanced by the need to remain politically distinctive. Being a list MSP allows me to represent areas where my party has no constituency MSP, yet members seek access to the party itself.
Another list MSP said the following, in a similar vein, in the 2003 survey:
The AMS list gives each elector a range of party parliamentarians to approach. I have been impressed by the significant number of people who have told me ‘this is the first time I have ever voted for someone who got in’.
It is clear that choice for voters has increased under MMP in Scotland and Wales, strongholds for the Labour Party. The big question is how Labour will deal with the unfamiliar competition that arises from this increase in choice for the voters that is facilitated by the new electoral system. The early indication is that Labour would like to reduce the competition. The Wales Office in a 2005 White Paper called for the abolition of dual candidacy in elections to the Welsh Assembly in response to the common situation where candidates who lose in constituency races enter the legislature via the party list. Citing what it called ‘considerable dissatisfaction that candidates who are rejected by a particular constituency can still become Assembly Members through the regional list and so be able to claim to represent the constituency that rejected them’, the White Paper says this situation of competition ‘can have done little to enhance the credibility of the electoral process or the voters’ level of engagement with it’.26
|
Conclusions |
|---|
|
TopThe constituency role in...Testing the electoral incentives...Competition over constituency...ConclusionsNotes |
|---|
In the process of their self-promotion, some list representatives in Scotland and Wales are stepping on the toes of their constituency-elected counterparts, many of whom resent what they consider to be an intrusion into ‘their patch’. The fact that most constituency representatives are from the Labour Party, historically dominant in Scotland and Wales and now in government there, only serves to enhance the perceived class distinction between these constituency-elected, government party representatives and the list-elected, opposition party representatives. The lack of public awareness of the presence of list-elected regional representatives and the difficulty for these representatives to cover large electoral regions have made competition in the form of shadowing by some list representatives quite common, while other list representatives focus more strongly on region-wide policies and deal extensively with interest groups.
Results from this research show that after two elections, Scottish and Welsh representatives elected in single-member constituencies spend more time helping constituents who have problems with government agencies than their list-elected counterparts, who spend more time in contact with interest groups. Those elected in constituencies also spend more time seeking public spending projects or promoting their areas as places for business. The 2003 survey of British representatives also reveals that constituency-elected Scottish Parliament and Welsh Assembly members are more likely than those elected from party lists to find constituency service to be very important to their re-election prospects. While these results are not conclusive, because of the limitations arising from the relatively small sample size and roughly 40% response rate, they are suggestive of an electoral incentive influencing the behaviour of elected representatives. Those representatives who are more interested in constituency service are, perhaps, better at performing it, and this may help them succeed at being re-elected. List representatives, however, sometimes compete strongly with constituency representatives over constituency service, particularly when they seek to win constituency seats.
The changes in MSP and AM survey results from 2000 to 2003 also indicate that some role differentiation, perhaps facilitated by the MMP electoral system, has taken place in Scotland and Wales. Those elected in constituencies are more likely to claim that they entered the legislature to serve their constituents, while list-elected representatives were more likely to say that helping to enact the policy preferences of their party into law motivated them to stand for election. In a rudimentary rational choice sense, list representatives may prefer to focus on party policy and interest group work, especially if they are likely to be near the top of their party lists, because re-election is very likely, and constituency service is an inefficient use of time and resources. Those list representatives who are not confident of high list rankings are more likely to compete heavily with constituency representatives over constituency service.
Does this apparent role differentiation mean that MMP creates ‘second-class’ representatives among list-elected members of the Scottish Parliament and the Welsh Assembly? If constituency members of these bodies can learn to accept a constituency presence by their list colleagues, then there should be no class distinction between the two types of representatives. In Germany, dual candidacy has been the norm for decades, and many of those who are defeated in constituencies enter the Bundestag or a Landtag via their party lists, setting up offices in the constituencies where they were defeated with no public outcry. Furthermore, dual candidacy is the norm among mixed-member electoral systems around the world. On the other hand, the high profile that many list members are creating for themselves is clearly frustrating to constituency members, most of whom are Labour Party members unaccustomed to having any competition at the local level in Scotland and Wales. Labour’s plan to abolish dual candidacy for Welsh Assembly elections, and the implication that their list-elected opponents are ‘second-class’ representatives, do not help the situation. Those elected from party lists may be elected in a way that differs from what the British are used to, but list representatives nevertheless play an important role in the politics of Scotland and Wales by adding pluralism where it was not abundant and introducing serious local competition where it did not exist before.
|
Notes |
|---|
|
TopThe constituency role in...Testing the electoral incentives...Competition over constituency...ConclusionsNotes |
|---|
I thank the following for their helpful advice: Anthony Messina, Michael Coppedge, Andrew Gould and Frances Hagopian, plus three anonymous reviewers. I am also grateful for support from the University of Notre Dame Department of Political Science, the University of Notre Dame London Centre and the School of Politics, Philosophy and International Relations at Keele University.
1 M.S. Shugart and M.P. Wattenberg, Mixed-Member Electoral Systems: The Best of Both Worlds?, Oxford University Press, 2000; A. Lijphart, Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries, Yale University Press, 1999; D.M. Farrell, Electoral Systems: A Comparative Introduction, Palgrave, 2001; A. Reynolds and B. Reilly, The International IDEA Handbook of Electoral System Design, International Institute for Democracy and Electoral Assistance, 1997. 
2 L.J. Ward, ‘Second-Class MPs? New Zealand’s Adaptation to Mixed-Member Parliamentary Representation’, Political Science, January 1998, p. 127.
3 S. Bowler and D.M. Farrell, ‘Legislator Shirking and Voter Monitoring: Impacts of European Parliament Electoral Systems upon Legislator-Voter Relationships’, Journal of Common Market Studies, March 1993.
4 B. Cain, J. Ferejohn and M. Fiorina, The Personal Vote: Constituency Service and Electoral Independence, Harvard University Press, 1987.
5 T. Lancaster and D. Patterson, ‘Comparative Pork Barrel Politics: Perceptions from the West German Bundestag’, Comparative Political Studies, July 1990; W. Patzelt, ‘What Can an Individual MP Do in German Parliamentary Politics?’, Journal of Legislative Studies, Winter 1999; H.D. Klingemann and B. Wessels, ‘The Political Consequences of Germany’s Mixed-Member System: Personalization at the Grass Roots?’ in M.S. Shugart and M.P. Wattenberg (eds), Mixed-Member Electoral Systems: The Best of Both Worlds?, Oxford University Press, 2000.
6 D. Searing, ‘The Role of the Good Constituency Member and the Practice of Representation in Great Britain’, Journal of Politics, June 1985 and Westminster’s World: Understanding Political Roles, Harvard University Press, 1994; P. Norton and D.M. Wood, Back from Westminster: British Members of Parliament and Their Constituents, University Press of Kentucky, 1993.
7 K.C. Wheare, Legislatures, Galaxy Press, 1963.
8 I. Crewe, ‘MPs and their Constituents in Britain: How Strong are the Links?’ in V. Bogdanor (ed.), Representatives of the People? Parliamentarians and Constituents in Western Democracies, Gower, 1985.
9 S. Weir and D. Beetham, Political Power and Democratic Control in Britain, Routledge, 1999, p. 70.
10 P. Norris, ‘The Twilight of Westminster? Electoral Reform and its Consequences’, Political Studies, December 2001, p. 891.
12 M. Kaase, ‘Personalized Proportional Representation: The "Model" of the West German Electoral System’ in A. Lijphart and B. Grofman (eds), Choosing an Electoral System: Issues and Alternatives, Praeger, 1984, p. 162.
14 C. Jeffery, ‘Electoral Reform: Learning from Germany’, The Political Quarterly, July–September 1998, p. 246.
15 T. Burkett, ‘The West German Deputy’ in V. Bogdanor (ed.), Representatives of the People? Parliamentarians and Constituents in Western Democracies, Gower, 1985, p. 130.
16 E. Jesse, ‘Split-voting in the Federal Republic of Germany: An Analysis of the Federal Elections from 1953 to 1987’, Electoral Studies, August 1988, p. 120.
17 W. Patzelt, ‘German MPs and Their Roles’ in W.C. Müller and T. Saalfeld (eds), Members of Parliament in Western Europe: Roles and Behaviour, Frank Cass, 1997, p. 60.
20 A. McCabe and J. McCormick, ‘Rethinking Representation: Some Evidence from the First Year’ in G. Hassan and C. Warhurst (eds), The New Scottish Politics: The First Year of the Scottish Parliament and Beyond, The Stationery Office, 2000.
21 T. Lundberg, ‘Putting a Human Face on Proportional Representation: Early Experiences in Scotland and Wales’, Representation, Spring 2002.
22 Lancaster and Patterson, 1990, p. 458.
24 W.M. Downs, Coalition Government, Subnational Style: Multiparty Politics in Europe’s Regional Parliaments, Ohio State University Press, 1998.
25 R. Donald, ‘MMP: Has it Delivered, Should it Continue?’, Representation, Spring 1998, p. 48.
26 Wales Office, ‘Better Governance for Wales’ (Cm. 6582), June 2005, p. 10.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||