As usual, we might look to other countries for insight and inspiration. It appears that Germany, Scotland, Wales and New Zealand all deal with list MPs somewhat differently. However, there are some broad roles that list MPs take on in these countries:
- One common practice is for list MPs to "shadow" their riding counterparts, serving as a second (or third, or fourth) contact point for constituents in some geographic area. In some places list MPs shadow a single riding; in others they take on larger regions.
- Some MPs explicitly represent concerns that do not fit neatly into a region. For example, some MPs in New Zealand take on the task of representing concerns of the Pacific Islander and Chinese minorities.
- A few list MPs are specifically recruited for that role because they are experts in some field. These MPs may be placed high on party lists and not contest ridings at all, and serve as policy advisors for their area upon election. From the literature I have read, this is the closest we see to ``party hacks'' getting ranked highly on party lists.
Sources: (Cody 2003, p. 40; McLeay Vowles 2007, p. 86)
In many ways all of the above countries consider list and riding MPs to be equivalent. For example, they sit together in legislature, their base pay is the same, they have equivalent voting power, and they all participate in committee work. (Massicotte Long 2004, p. 61) (Bradbury Mitchell 2007, p. 119) The differences have to do with the way list MPs are perceived by the public, the duties they are (implicitly or explicitly) expected to carry out, and the relationships and tensions between list and riding MPs. In these areas the political cultures of different countries differ quite a bit.
By most accounts, the Germans accept their list MPs most readily, and the roles of list and riding MPs differ the least. "Shadowing" is both prevalant and widely accepted -- citizens tend to take their constituency work to MPs based upon party affiliation and values as much as geographic considerations. (Massicotte Long 2004 p. 61-62, Lundberg Perceptions 2006 p. 64)
A paper by Patzelt (Patzelt 2007) classifies list MPs as those from small parties, those who wanted to win ridings but failed, and those who were "unexpected" list MPs -- people who did not contest ridings, but ended up in power despite being low on the list. He suggests that the incentives to shadow depend on the classification of list MP: most MPs from small parties pose no real threat in winning ridings. Although they do some constituency work they focus their attention on special interests and pleasing their party elite. The other two classes of list MPs invoke more conflict from their riding counterparts. Would-be riding MPs want to contest (and win) ridings in subsequent elections. "Unexpected" MPs realize that their status as riding MPs are vulnerable because of their low rankings. Thus these groups work harder at being visible to the voters. (ibid, p. 65-66) Both these groups threaten the dominance of their riding counterparts, but the riding MPs accept this as "part of the game". Interestingly, Patzelt does not include list MPs selected for their expertise as a category in Germany; list members are either from small parties or somewhat uncomfortable being on the list.
It is not clear that this multi-fronted outreach does a lot of good in Germany, however. Patzelt also publishes a scary table (ibid, p. 63) that indicates that 45% of Germans couldn't name a member of their federal parliament, and 73% of respondents felt it was "rather difficult" or "nearly impossible" to get to talk with a member of their federal parliament! (It is not clear whether this included East German respondents; in most of this paper Patzelt focuses on the Western German politicians only.) Lundberg found that overall both riding and list members of the German parliament placed less importance on constituency work than their Wales and Scottish counterparts. (Lundberg Perceptions 2006, p. 71)
In terms of workload distribution, it appears that list and riding MPs do fairly similar work. Patzelt reports that riding MPs (both federally and in the Lnder) tend to favour constituency work and local activities more than their list counterparts, but the differences are small. Similarly, a paper by Stratman and Baur (Stratman Baur 2002) find some weak evidence that federal German riding MPs tend to join committees that benefit their local constituents, while list MPs tend to join committees that are affiliated with special interest groups that are not focussed geographically (such as Family/Child/Elderly matters). The time estimates in Patzelt's paper do seem to suggest that list MPs do less work than riding MPs overall, however.
It is interesting to note that riding MPs tend to have more status than their list counterparts. For example, if list members and riding members have the same affiliation, the riding members are considered the "boss" (Patzelt 2007, p. 66). It is also the case that list MPs shadow their riding counterparts -- not only because the parties expect this behaviour, but because they want to contest riding seats.
It is not immediately obvious why this should be the case in Germany. Firstly, there is a 50-50 split between riding and list seats. Secondly, Germany used a pure list system in the Weimar Republic, so the country has some experience with list MPs. Nonetheless, riding MPs feel decidedly more secure about their chances for re-election than list MPs (ibid, p. 51). My guess is that incumbent bias might remain a factor, although I don't think it explains everything.
Massicotte (Massicotte Long, p. 61-67) goes through a detailed analysis of list vs. riding MPs, concentrating mostly in Germany. He notes that list and riding MPs in Germany get equivalent pay and allowances. He finds that more heads of state and cabinet ministers tend to be riding MPs, but that is because the big-tent winning party tends to win a lot of ridings. He takes the view that there are no major differences between list and riding MPs in Germany.
Wales and Scotland
Of the countries I have been focussing on in this series, list MPs have had the roughest time in Scotland and Wales. Most of the literature I have read claims that tensions run high between riding MPs and their list counterparts. (Lundberg Competition 2006, Bradbury Mitchell 2007) This is especially interesting in the case of Scotland, which is the only country that provides explicit guidelines for the duties of its list members. Scottish List MPs are supposed to distinguish themselves clearly from riding MPs, and they are supposed to serve regional concerns. (Bradbury Mitchell 2007, page 119) In both Scotland and Wales, regions are well-defined by the voting system. Each country is divided into regions, and party votes are tallied regionally, with no considerations for overall proportionality. This results in some vote distortions (especially in Wales, which has five regions and twenty list members in total) but it gives list MPs a strong geographical focus.
Scottish guidelines do not define "regional concerns" very tightly other than to say that list MPs should be active in more than one riding. In practice this does not always happen; many list MPs will either focus on their "home" riding or a riding where they think they could win the next constituency race. This was especially prevalant in the Conservative party, partially because list placement in that party depends on riding performance. However, Plaid Cyrmu members (from Wales) and members of both the Scottish Nationalists and the Liberal Democrats (in Wales) also admitted to shadowing with the hopes of contesting future nominations. (ibid, p. 131) Given the gains made by Plaid Cyrmu and the Scottish Nationalists in the 2007 elections, my guess (which I have not analysed at all) is that it may have worked.
Bradbury and Mitchell found that Scottish list MPs were slightly more likely to deal with groups and organizations rather than individual constituents; overall attention was balanced except in the case of the Scottish Green party, which focussed on groups and organizations (ibid, p. 133)
In contrast, Lundberg (Lundberg Perception 2006, p. 66) found that riding members tended to spend more time than list members helping constituents, and that this gap grew from 2000 to 2003. In 2000 riding MPs who responded to Lundberg's survey claimed to spend 17.5% of their time helping constituents, as opposed to 15.4% estimated by their list counterparts. In 2003 the estimates were 21.1% of riding MP time, vs. 13.7% from list members. Although I am skeptical of the numbers themselves (they are self-reported and reported at too high a precision) it seems clear that something changed between 2000 and 2003. If nothing else, MPs in 2003 had changed their perceptions of how much constituency work they needed to do in order to be considered respectable.
Some of the reaction to MMP in Scotland and Wales is influenced by the circumstances under which MMP was adopted. Unlike New Zealand and Ontario, voters did not vote in referenda dedicated to the adoption of a new voting system; rather, MMP was part of a larger package called "devolution". Devolution was intended to create distinct parliaments for Scotland and Wales, thus giving them somewhat more autonomy over their politics (and appeasing Scottish and Wales nationalists in the process). In Scotland the referendum for devolution passed with 74.3% support; in Wales the referendum squeaked by with 50.3% support. (Alvarez-Rivera 2007) Devolution resulted in the creation of two new assemblies: the Scottish Parliament (SP) and the National Assembly of Wales (NAW).
Because MMP was part of a larger package, it is not clear that the Scots and Welsh wanted a new voting system as much as they wanted some additional power; in particular it looks like the Welsh are not so enthusiastic about having MMP imposed on them -- voter turnout in Welsh elections are abysmal. (Electoral Commission Wales Facts 2007, p. 4)
Another big factor for Scotland and Wales has to do with the Labour Party, which tends to dominate politics in both these regions. Until the 2007 elections, the Labour Party won the vast majority of ridings (and almost no list seats) in both Scotland and Wales. This has given the riding seat/list seat controversy a distinctly partisan spin, and the Labour Party has done whatever it can to make life hard for list MPs. The Scottish guidelines distinguishing riding and list member roles is one manifestation of this. Another is that list members in Scotland get smaller office allowances, supposedly because list members are expected to share "regional" office space with each other. (Bradbury Mitchell p. 119) In practice, list members complain that the smaller allowances mean they can do less constituency work, which suits riding members just fine.
The practice of shadowing annoyed the Labour Party in Wales so much that they attempted to ban dual-candidacy in 2004, and finally succeeded in 2006. This means that politicians in Wales have to decide whether they will run for a riding (possibly challenging the Labour incumbent) or whether they will opt for the greater job security of the list. In my view this move is a pretty blatant sign that riding members in Wales were feeling the heat. Overall, dual candidacy is widely permitted in almost all MMP and parallel systems: according to Lundberg (Lundberg Competition 2006, p. 113) the only places that ban it are the Ukraine, Thailand and the Palestinian Council. In addition, Mexico limits the number of dual-candidates to 60 per party.
One negative side to shadowing expressed by MPs is that constituency work becomes politicized. List and riding members fight for the more glamourous constituency cases, and citizens choose the MPs that share their ideological views in the hopes of getting a more representative audience. This erodes the perception of constituency work being nonpartisan, where a Conservative MP will work just as hard on behalf of an NDP voter as a fellow Conservative. (Lundberg 2006 Perceptions, p. 72-74)
I have not been able to find a lot of public opinion studies on the support of list members in Wales and Scotland. Some focus group work conducted in Wales indicates that knowledge of electoral systems there is pretty low. Only five of 70 focus group participants identified their MMP system (called the "Additional Member System" or AMS there) by name. (Electoral Commission 2006 Wales, p. 41) Although focus group participants knew that they were represented by multiple MPs at the National Assembly of Wales level, "only a minority" of participants knew there were two types of MP (ibid, p. 35). Similarly, the issue of dual-candidacy barely registered with focus group participants, coming up in only two of the eight focus groups (ibid, p. 44), and not receiving much attention even there. As a whole, Welsh focus group members did not feel much connection to any of their National Assembly of Wales MPs, and felt that on the whole the Assembly was ineffective. (ibid, p. 31, 35) Given these perceptions of irrelevance and overall ignorance of MMP mechanics, it is perhaps not surprising that there are no large public outcries against list members. What is surprising is the huge disconnect between the strong feelings for list MPs among the political elite compared to the incredible disinterest in the issue among the Welsh voting population.
Although the Scottish people appear to have a better opinion of devolution, a lot of confusion about the differences between list and riding MPs remain. The Arbuthnott Commission studied Scottish boundaries and voting systems. In their 2006 report, they indicated that instead of tensions between list and riding Members of Scottish Parliament (MSPs), there was more confusion between the roles of Westminister MPs and MSPs, and that people tended to take their local concerns to their local council members and their bigger concerns to their Westminister MPs. (Arbuthnott 2006, p. 59) As in Wales, few people knew their regional MSPs (Electoral Commission Scotland 2006, As in Ontario, it is clear that people do not like closed lists very much: in 2003 surveys, about 45% of people disagreed that parties should be able to choose the lists, and about 40% agreed that some form of open list system would be better. The Arbuthnott Commission ended up recommending open lists, although I don't think this recommendation was carried out. (ibid, p. 33, 45).
Overall it appears that Scots value their Scottish Parliament more than the Welsh do theirs. More people know that the Scottish Parliament exists, and they feel that the parliament represents them and is more approachable than Westminister MPs. (Electoral Commission Scotland 2006, p. 22, 23, 32) But there is so much confusion between the many different levels of government in Scotland that not too many regular people seem to care about the differences between list and riding MSPs.
It appears that public perceptions of list MPs has been changing as New Zealand adjusts to MMP. Early perceptions of list MPs were rather negative, in part because a list MP (Alamein Kopu) decided to quit the Alliance Party and sit as an independent (LCC 2004, p. 157, footnote 22), This combined with the New Zealand First debacle of 1996 gave MMP in general and list MPs in particular a black eye. (Vowles Banducci Karp 2006, p. 41) In polling done for the 2001 Review of MMP in New Zealand, 61% agreed that list MPs are not as accountable to voters as electorate MPs, and only 15% disagreed. Associated with this were some calls for open lists, although the Review Committee did not recommend this. (MMP Review Committee 2001, p. 85)
List MPs have continued to be perceived more poorly than their riding counterparts, although neither class of MP earns a lot of popular support under MMP. In 1999 17% of those polled for the New Zealand Election Study stated that they approved of riding MPs in general, compared to just 6% who approved of list MPs in general. By 2002 this had gone up to 22% who approved of riding MPs in general, compared to 10% who approved of list MPs. There is an important caveat to these dismal results, however: people's approvals of specific MPs (especially MPs they have had contact with) is much higher. In 2002, those who recalled the name of their riding MP supported those MPs at a rate of 56%; those who recalled the name of a specific list MPs gave 49% approval of that list MP. Those who actually had contacted their MPs was even higher: in 2002 71% of people who contacted their riding MP approved of him or her, and 75% of those who contacted a specific list MP approved of him or her. (McLeay Vowles 2007, p. 84)
What message can we draw from these strange results? First of all, those who are politically aware enough to be able to name specific MPs tend to have higher approval of MPs in general, and people who are not politically engaged look down upon all MPs. Those who contact list MPs probably have some sympathy for those list MPs to begin with, so maybe it is not surprising that the (many fewer) people who approach list MPs for help tend to appreciate those list MPs more.
New Zealand does not formally mandate roles for its list MPs. Quite a bit of shadowing and local representation does occur, although the number of list MPs opening local offices declined between 1999 and 2003. In 1999 every list MP was assigned to shadow one or more ridings (some MPs had up to nine ridings to look after!).
It appears that shadowing went down as the years passed. By 2003 only 74% of list MPs had opened some kind of constituency office, with some list MPs sharing offices with their riding counterparts. (McLeay and Vowles 2007, p. 82) As in Scotland, list MPs receive less money for local offices than their riding counterparts get, which again limits the amount of constituency work that list MPs can take on. (ibid p. 74). Unlike Scotland and Wales (but like Ontario) the voting system does not create natural "regions" that list members can claim as their own. McLeay and Vowles found that members of smaller parties had fewer hopes for winning riding seats, and thus had less incentive to focus on casework. In contrast, list MPs from bigger parties (which had more realistic hope of winning ridings) still saw constituency work as important. (ibid p. 87) There is evidence that some list MPs -- particularly those from ethnic minorities -- took on the role of serving as ambassador for that ethnic group (ibid, p. 89)
Overall, McLeay and Vowles found that riding MPs tended to do more casework than list MPs, and that the amount of casework riding MPs took on went way up after MMP was introduced. There are two explanations for this: after MMP the number of ridings was reduced, and most people chose to take their concerns to riding MPs over list ones. (ibid p. 85)
One of the more intriguing consequences of list MPs comes from a paper by Heitshusen, Young and Wood. They conducted surveys of MPs from six different houses: MPs from Canadian legislature and British House of Commons (elected using FPTP); the Irish Dail and Australian Senate (elected using variations of Single Transferable Vote, a proportional system); the Australian lower house (elected using Alternative Ballot, a non-proportional system); and the New Zealand legislature (elected using MMP). They asked members of each of these houses to assign priorities to a set of job responsibilities. They then determined how highly each class of MP valued constituency work as compared to their other duties. This helps correct for self-reporting inflation.
The results are striking. A majority of constituency-based MPs (in Canada, New Zealand riding MPs, the Australian lower house, and Britain) ranked constituency work higher than their other responsibilities. In contrast, only 1 of the 30 New Zealand list MPs interviewed did, and no Australian senator did at all. 71% of the New Zealand riding MPs ranked constituency work as their highest concern -- much higher Canadian MPs, which finished second with 57% of MPs rating constituency work highest. (Heitshusen Young Wood 2005, p. 39) McLeay and Vowles argue that some of this is due to New Zealand culture, which places a high value on constituency work. That alone does not explain the behaviour of list MPs, however. Heitshusen, Young and Wood offer four other hypotheses (ibid, p. 40):
- riding sizes went way up after MMP was introduced, so riding MPs had to put in more effort to deal with the extra constituency work.
- MMP fragmented the party system and made fewer seats "safe seats".
- List MPs shadowed riding ones, making the riding MPs feel less secure in their jobs.
- New Zealand ridings tend to be more rural, and for some reason rural ridings foster closer MP-constituent relationships than urban ones do.
Which of these hypotheses is correct? It is hard to say. Evidence suggests that MMP actually makes ridings safer than they are under FPTP, because people can vote for incumbent they like independent of party affiliation. My guess is that MMP has something to do with this dramatic jump in the priority of constituency work for riding MPs. Even if list MPs themselves do not work as hard in ridings as they should, their presence may encourage their riding counterparts to work harder. The downside to this is that one of the basic criticisms of FPTP defenders is correct: it does appear that there are some important differences between being a list and riding MP in New Zealand.
Conclusions and Predictions
Predicting the future is dangerous business, especially in political science. There are too few countries and too many variables to foresee consequences with a lot of confidence.
The job of trying to predict the roles of list members in Ontario is even harder than usual, because perceptions and roles for these MPs vary widely between the countries under examination. Nonetheless, I tentatively offer the following predictions.
It appears that to some extent shadowing of riding MPs happens everywhere. The tradition is strongest in Germany, and tends to be more pronounced in situations where the list MPs hope to become riding MPs one day. Because Ontario uses a province-wide list for list MPs, those MPs do not have natural regional boundaries in which to concentrate their efforts. Thus, I expect that although there would be some shadowing in Ontario, it would be practiced mostly by members of the bigger parties.
I do see some list MPs (particularly in the Greens and other ideologically-focused parties) acting as ambassadors for their causes. I could definitely see this happening in Northern Ontario. I do not forsee as much ethnic advocacy as we see in New Zealand; their history with respect to minorities and especially aboriginal relations is quite different from the Canadian experience. Unlike other MMP advocates, I doubt that we will develop the cultural convention of assigning list MPs to regional issues.
I think that we would avoid the problems Scotland and Wales have of a partisan split between riding and list seats, because we have at least two parties in Ontario that can win both ridings and list seats.
I do think that list members would suffer from a poor public image in Ontario. If nothing else, so much of the current referendum campaign has focused on the deficiencies of list MPs that the deck would be stacked against them right from the start. I think that list MPs would attempt to be visible in the public eye (via constituency work and photo ops) but nonetheless most people would not be aware of the list representatives assigned to shadow their ridings. As in New Zealand, I think disapproval for list MPs would be high except among those who seek out interactions with their list MPs.
I think there could be a lot of confusion over the proper roles list MPs are supposed to play, and that this would confuse and frustrate both list and riding MPs during the first few years. However, I do not think that list MPs will be quite as lazy (or as sycophantic) as FPTP defenders would have you believe. Certainly in legislature work they will work just as hard as riding MPs, and although they will have fewer constituency duties I expect most of them will have some presence in their communities, if only to demonstrate that they (and their associated party) deserve re-election.
(Alvarez-Rivera 2007) Manuel Alvarez-Rivera. "Election Resources on
the Internet: Elections to the Scottish Parliament and the National
Assembly of Wales". http://electionresources.org/uk/sct-wls/
May 15 2007, accessed July 31 2007.
(Arbuthnott 2006) Commission on Boundary Differences and Voting
Systems (chair: Sir John Arbuthnott). Putting Citizens First:
Boundaries, Voting and Representation in Scotland. Edinburgh: The
Stationery Office. ISBN 0-10-888179-2. Available from http://www.arbuthnottcommission.gov.uk/FinalReport.htm
(Bradbury Mitchell 2007)
Jonathan Bradbury and James Mitchell. "The Constituency Work of
Members of the Scottish Parliament and National Assembly for Wales:
Approaches, Relationships and Rules.
Regional and Federal Studies, vol 17, no 1, 2007, pp. 117-145.
(Cody 2003) Howard Cody. "Early Lessons from Mixed-Member
Proportionality in New Zealand's Westminister Politics", The New
England Journal of Political Science, Vol 1, No. 1, 2003, pp.
(Electoral Commission Wales 2007) The Electoral Commission. The
National Assembly for Wales elections 2007: Facts and Figures.
Available from http://www.electoralcommission.org.uk"
(Electoral Commission Scotland 2006) The Electoral Commission.
Scotland -- Poll Position: Public attitudes towards Scottish
Parliamentary and local government elections. ISBN 1-904363-83-0.
Available from http://www.electoralcommission.org.uk"
(LCC 2004): Law Commission of Canada, Voting Counts: Electoral
Reform for Canada, Queen's Printer of Canada (2004). ISBN
J31-61/2004E (and available online).
(Lundberg Competition 2006) Thomas Carl Lundberg. "Competition
between Members of the Scottish Parliament and the Welsh Assembly:
Problem or Virtue?", The Political Quarterly, vol 77, no 1,
January-March 2006. p. 107-117.
(Lundberg Perceptions 2006) Thomas Carl Lundberg. "Second-Class
Representatives? Mixed-Member Proportional Representation in
Britain.", Parliamentary Affairs, Vol 59, No 1, 2006, pp.
(Massicotte Long 2004) Louis Massicotte. In Search of a
Compensatory Mixed Electoral System for Qubec,
Gouvernment du Qubec, 2004. ISBN 2-550-43379-3.
(McLeay Vowles 2007) Elizabeth McLeay and Jack Vowles. "Redefining
Constituency Representation: The Roles of New Zealand MPs under MMP",
Regional and Federal Studies, vol 17, no 1, 2007, pp. 71-95.
DOI: 10.1080/13597560701189628 .
(MMP Review Committee 2001) Rt. Hon Jonathan Hunt, chair. Inquiry into the
Review of MMP: Report of the MMP Review Committee, New Zealand
House of Representatives, August 2001.
(Patzelt 2007) Werner J. Patzelt. The Constituency Roles of MPs at the
Federal and Lnder Levels in Germany",
Regional and Federal Studies, vol 17, no 1, 2007, pp. 47-70.
DOI: 10.1080/13597560701189610 .
(Stratmann Baur 2002) Thomas Stratmann, Martin Baur. "Plurality Rule,
Proportional Representation, and the German Bundestag." Center for
Economic Studies and Ifo Institute for Economic Research, Working
Paper Number 650 (2). January 2002. Available from http://papers.ssm.com/abstract=301069
(Vowles Banducci Karp 2006) Jack Vowles, Susan A. Banducci, Jefferey
A. Karp. "Forecasting and Evaluating the Consequences of Electoral
Change in New Zealand", Acta Politica</a>, vol 41, 2006, pp.