19th May 1999

(3,524 words)

"There is no majority party in United States electoral politics". Discuss, clarifying the implications for governing.

When discussing American electoral politics , some key themes should be addressed. Understanding of these and the reasons behind them will allow for further evaluation of if it is possible to talk of a 'majority party' in the American context. These themes include the moves towards candidate centred campaigning, voter dealignment and split-party voting. In addition to these factors are those which date back further. These include organisation of parties predominantly at a state, rather than a national level. Also, the constitutional arrangements of the separation of powers have implications on the political parties.

When these are addressed, I believe it suggests that there is indeed little justification of claiming there to be a majority party in the United States. This at first seems surprising because in a two party system, the party with the most elected candidates seems to hold a majority. The problem is that this is a superficial assessment and majorities cannot be counted simply by bums on seats. In relation to 'electoral politics' it is important to see whether those candidates were elected because of their party affiliations or for other reasons. Only if it is accepted that the party was the base of their electoral support can a party claim a majority. Further, not only must the party be the basis of candidate's success, but it must be, for example in House elections, the national party that allows a candidate victory. The point being made is that if a representative is elected as a Democrat in California and another is elected as a Democrat in Nebraska, there must be more than superficial links between the Democratic parties in California and Nebraska for the Democrats to claim to have a majority. Polsby suggests that the US has "a hundred parties under two banners" . If he is correct, then even in when a party has the most seats, they should not be seen as a majority party. This observation will of course have major implications for governing. A majority of seats is a paper tiger unless those members can be persuaded to act in a partisan fashion. The general inability of the American system to organise partisan behaviour in Congress should be seen as an argument against the existence of a majority party.

Those that accept a congressional majority as a case of a 'majority party' could suggest that another form of 'majority party' can be seen when the presidency and the legislature are both held by the same party. This again, superficially, seems correct. If after a presidential election, both Congress and the White House are held by one party, it does appear that the electorate have given support to that party. This is, however, not necessarily the case. The reasons for this will be the basis of this study. Since the war, divided government has been the norm. Under these conditions it might be suggested that a majority party only occurs under united government. This, however, seems false. Instead, I wish to suggest that united government is often coincidence rather than an indication of a 'majority party'. The different branches are often elected on completely different issues. Neither branch's electoral support is necessarily based around their parties.

A 'majority party' cannot therefore be measured simply by electoral returns. A majority party should be an organisation with a consistent (or at least not self-contradictory) platform across those elected as part of the party. The lack of cohesiveness within American parties makes such a possibility remote.

The reason for the lack of cohesion in the national parties stems largely from their evolution. There are two aspects of this worth highlighting. These are the entrenched natures of the parties as 'public utilities' and the emphasis of state, rather than national organisation. The notion of parties as 'public utilities' comes from Leon Epstein . In the American system, parties are seen a vehicles for public participation in democracy, rather than the European model of parties as private organisations competing to be part of the democracy. American parties are not mass membership organisations with joining fees and centralised hierarchies who expel fringes whose views they find out of line. Instead, American parties are open to all and with loose organising structures. In many states, to be seen as a member of a party, you must simply mark a box on the voter registration form. This often gives you a right to select candidates through primary elections. The fact that such a variety of opinions are therefore given a voice within the two American parties shows that even in small regions there is a difficulty in mounting a coherent party issues platform. These problems would be assumed to magnify greatly when looked at on a national scale.

As was hinted at above, party structures are largely dictated by state, rather than national legislation. Epstein notes that little, except some campaign finance legislation in the 1970s, has been done to nationally regulate parties. The result of this is a vast diversity of regulations between states. This leads to differences in party structures across the country. Before even noting differences in regional interests and issues, the difference in party structures can have an impact on the national party. Primaries will have different franchises between states and varying abilities of cliques to control parties, depending on state regulations. This makes it hard to develop a national strategy or national party structure. The only time that all the state parties seem to be organised together are at the national conventions. These are far from business like meetings, so preventing an image of American parties as having a national policy stance or structure.

In addition to these structural factors, differences between interests and issue cleavages between state will be expected to prevent the creation of coherent national party policy platforms. If the ideological centre point in one state is to the right of that in another state, it is far from inconceivable that members standing for different parties in different states will have similar views. Before the realignment of the South (since the 1960's), this was certainly the case with Democrats being seen as on the ideological right of Republicans from northern states. This must be seen as being related to having only two parties representing the views of over two hundred million people (Israel's population of less than five million manages to support fourteen parties). Variation of issues between areas, characterised by such notions of the southern 'bible-belt' of conservative voters, compared to the cosmopolitan liberal stereotypes of major northern cities highlights this. For the two parties to be competitive in all districts, they must contain a variety of campaigning platforms within each party. If parties had clear national agendas across the country, they would simply be surrendering districts to their opposition. With the failure of third parties to operate in the American system, such surrendering would remove electoral choice other than in the dominant party's primary elections.

With this theory in mind, it is now possible to look at the practicalities of electoral politics in the United States. As was suggested in the introduction, there are some features of elections which seem special to the United States. The first of these is the candidate centred campaign. The impact of this form of campaigning on party majorities will be strong. If people are being elected as is individuals, rather than as representatives of a party platform, it does strongly suggest that parties can claim no majority. This form of campaigning has increased in recent years and at all levels, as well as for various reasons. With respect to governing this will also have implications. A candidate elected on his own merits, rather than those of the party will have limited to follow the party whip as might be expected in other parliamentary systems.

In presidential elections, the disparity of views within a party across the country means that to only try to appeal to those that identify with your party would cut out a large proportion of the electorate. These include those who identify locally with an opponent's party, but would still be open to a particular candidacy. The freeing from party ties allows a presidential candidate to personify issues to move to positions on cleavages which otherwise the party would not adopt. An example of this was seen by Clinton's adoption of more conservative policies during the 1992 election. This point is highlighted when voters are asked about how they view the partisan nature of presidential candidates during the campaigns. This trend is moving further away from partisanship. In 1952, 9.2% of the electorate expressed a like or dislike of Eisenhower because he was a Republican. This contrasts with 1980 when only 2.9% mentioned Carter's party affiliation and 4.1% mentioned Reagan's . The lack of partisanship in the presidency is seen in cabinet appointments. For example, the current Defence Secretary, William Cohen is a former Republican.

Third candidates in presidential elections, such as George Wallace in 1968 and particularly Ross Perot in 1992 also point to the increased candidate centred nature of presidential elections. Although these did claim to head parties, their campaigns were centred around the individuals. Success of such campaigning (Perot and Wallace each gained 14% of the popular vote) perhaps show that the electorate are electing a man to do the job of chief executive, not the party that is beneath.

Candidate centred campaigning is not, however, confined to presidential elections. Congressional candidates may also wish to disassociate themselves with parties to appeal to a broader electorate. Perhaps there is a reverse affect from that noted for presidential candidates. Whilst a presidential hopeful might want to distance himself from regional strands of the party, those campaigning in these regions might want to distance themselves from those in Washington. Whilst it is clear that no one could claim to represent the national Democratic party, or the national Republican party, plenty of people in Washington try, because of their positions in federal institutions. This is typically seen to be the presidential candidates (in presidential years) or the various congressional leaders and the president in mid-term elections. Although they should not be seen as the leaders of the national party, this is often how they present themselves, the media presents them and so the voters see them. If a congressional candidate doesn't want to be associated with their views, they would be more inclined to centre the campaign around themselves.

Also, candidates have a large amount of personal prestige in their districts due to their personal activity, regardless of parties. This is highlighted by incumbency return rates of astronomic proportions, over 98% at the 1998 mid-terms. The size of the offices of congressional members in their home districts adds to the perception of their personal activity, distinct from what their national party is doing. Even if this is not personally centred, such as if activity is seen to be done by the local, rather than national party, it questions whether the national party can claim to be a 'majority party' when it wins most of these seats.

Independent candidates have also be seen at non-federal level. The most recent example of this is Jesse (the body) Ventura became the Governor of Minnesota in 1998. At lower levels still, it is often not hard to understand why many voters do not worry themselves too much about the party affiliations of those they elect as coroner or dog warden.

There is, however, now the need for a big 'but'. The above discussion suggests that there is an increasing emphasis on candidate campaigning, but, candidates still stand for parties and the frequently talked of decline in American parties has not removed them from their entrenched positions in the political system. There can be no argument that parties still have an impact. If there was true candidate centred politics, there would not be the clear regional party strongholds evident by looking at any electoral map. Regional perceptions are entrenched in the American system, and although can be shifted by realignments (such as that in response to civil rights legislation and migration to the south since the 1960s), the American voter can be seen to identify with a party. Angus Campbell's work, 'The American Voter' , highlights a model based about research in Michigan which showed that there is active party loyalty evident by individuals. This can vary over time, but the model suggests that the best predictor of how a voter will vote in the next election will be to analyse their party voting record. Whilst many recent studies have pointed towards moves to more volatile voting among sections of the electorate , the Michigan model does point to a reservoir of partisan votes that could be tapped by a candidate who stands under a party banner. What is not, however, clear is whether these party affiliations are to the national or local parties. If they are not to the national parties, partisan votes cannot be claimed to help build a 'majority party' in Congress or in the federal government.

There are other reasons for candidates to centre their campaigns on themselves and still stand under the party banner. The first of these is the resources available through parties. This is not simply in capital, but also the mobilisation of grassroots support for campaigning can tap into party allegiances. Whilst the media age of television and mail-shots does diminish the need for such an army somewhat, campaigning is still seen as a labour intensive procedure. Candidate's party affiliations also make life easier for the voter. This is related to models shown in Michigan research. For an electorate uninterested in politics and not keen to seek out information on candidates, knowing what party candidates are from gives a good benchmark to decide where to vote. Again, this does not necessarily mean that the voter is voting for a national party so even this form of voting behaviour does not entitle leaders in Washington to have a 'majority party'.

Another important feature that is central to modern American electoral politics is split-ticket voting. As was noted above, the post-war norm of formation of the federal institutions is to have different parties with electoral advantage in each. Even when on the same ballot, presidential and congressional candidates of the same parties often do not get elected together. For example, Nixon gained 61.8% of the popular vote (96.7% of the electoral college vote) , yet still faced a Democratic congress. One explanation is that voters believe that divided government is a good thing for the checks and balances of separated government and so purposefully split their vote. This, however, is often dismissed. A more likely explanation is simply that party affiliation is not the core reason for voting decisions. This ties in with the discussion above about candidate centred electioneering and the powerful position of incumbency. The implications of this towards the question is that again it can be suggested that no party can claim a majority. This should be contrasted with a European state such as Germany or the UK where voters are seen to be voting for national party platforms, so allowing leaders to claim to be 'majority parties'.

My conclusion that there is no national 'majority party' would of course be disputed by many. One who would certainly complain would be Newt Gingrich. Through his 'Contract with America' in 1994 he sought to produce a policy platform for the national GOP to campaign on. Further, his congressional victory in the mid-term elections that year, seem to suggest that the voters liked this platform. This was the first time the GOP had a majority in the House since Eisenhower was president. According to Ware , however, whilst some voters might have applauded this move, most had never heard of the Contract. The proof of a party majority is, however, seen in how cohesive the national party acts. In the early stages of the Contract there was certainly unity amongst GOP members of the House. Of the 73 GOP freshmen, 53 voted with the party 100% of the time in the first 100 days of the 104th Congress . This, perhaps, does show that whilst the Contract did not allow for a majority party, the realignment of the South may have. Although the GOP has regained control of congressional seats since then, power within the party structure has diminished strongly and so such party cohesion in congressional action has been seen less. An exception to this was the case of the Clinton impeachment where partisanship was evident in the House vote. Interestingly, psephologists have suggested that the Clinton scandal had little impact on the 1998 mid-term elections. A conclusion from this might be that although Representatives thought they had a mandate to behave nationally in a partisan manner, the electorate disagreed and voted on more traditional issues.

The Gingrich experience shows that there can be great implications for governing if it is perceived that representatives are part of a majority party platform. Even under Gingrich this majority was not as well disciplined as would be expected by European models. Fewer than 75% of legislation was voted on in a partisan manner . This figure alone undermines a notion of a 'majority party'. A point that is also made is that party cohesion in Congress comes in cycles, with the last fifteen years demonstrating a time when cohesion in strong. This contrasts with periods in the 1970's when partisan voting was less than 40%. Before this, in the 1950's partisan voting was at the high levels seen in the late 1980's and 1990's . Ware affirms this cyclical view, suggesting that such cohesion should be expected to reduce in the future.

The last area to address is the implications of governing in a separated system without majority parties. When both parties hold a branch of government, they can both claim a 'party majority' and so fight for the application of their perceived mandates. A more interesting occurrence is when both executive and legislative branch are held by the same party, yet there is disagreement between them. This highlights the difficulty in having a 'majority party'. Both would claim to represent to platform of the 'majority party', yet there will be two platforms and only one party. It is impossible for them both to represent this 'majority party'. The solution is to use the arguments above to suggest that neither is right and there is no majority party. Carter's presidency was a clear example of this. He was a Democratic president with a small electoral majority facing a Democratic House with a larger electoral majority. Both had been elected in the same set of elections and both believed that they represented the Democratic mandate to govern. When the two bodies therefore clashed, it caused major difficulties for the president. His ability to gain the 1980 Democratic nomination, does however suggest that it is not fair to suggest that he was no longer accepted by Democrats. Equally, however, the ability of the congressional Democrats to retain their majority in the 1980 elections suggested that it would not be fair to suggest that they were no longer accepted by Democrats (they won as Reagan took the White House). In a separated system, this must show that unless there is unity between the agenda of president and congressional leaders, there can be no 'majority party'. In addition to this, research by David Mayhew suggests that there is no difference in the rates of legislative productivity between periods of united and split-party government. This further undermines the role of parties as the basis for political rule in the United States.

In all, whilst superficially parties do appear to hold majorities, when looked at in greater depth, there is a suggestion that they do not. If the notion of 'majority party' is used in comparison to parties in other systems, it is clear that the American parties lack the internal structures to produce the uniformity seen elsewhere. This lack of internal national structure means that parties are unable to exhibit unity in governing. Also, this leads to variation in the party across the country, so giving different mandates to the same party in different regions. In addition, the complication of separated government and candidate focused electioneering undermines any sense of a centralised parties. Some examples, such as Gingrich's early successes do seem to contradict this slightly, but these are exceptions which are seen to be short lived. The implication for governing is that not only does it make it difficult for party leaders to discipline their members, but it is also not clear what policies can be suggested to have an electoral mandate.

The last, slightly flippant, point to make is that of turnout. In any election that fewer than sixty percent of those eligible to vote, do so, it is hard to justify any suggestion of a group holding a 'majority' within a country. This is consistently the case in the United States, with mid-term elections often managing turnouts of less than forty percent.