Account for the differences in the party systems of the UK,
the USA and Germany.

In comparison to other party systems around the world, those of the UK, the USA and Germany seem to exhibit many similar traits. These will be explored later, yet for all of the similarities, there are also many differences. The form of this study will be to first highlight some of the major differences between the three systems, followed by an attempt to account for them. This explanation will try and come from three directions. Firstly, I will look at two theoretical approaches to party theory. These are Duverger's linkage between party systems and electoral system type and the Downs/Sartori linkage between variation in ideology amongst the electorate and the party system. The second approach will look at the historical evolution of the party systems. The third angle will be looking at the institutional and social features (beyond electoral systems) of each country and see if these explain the differences highlighted. I will then try and conclude by seeing which approach, if any, best explains the differences between the states.

In order to highlight the difference, various aspects of the party systems shall be compared. The first of these related to the actual number of parties in the system. This measure, as Sartori points out is hard, because even in a classic two-party example such as the UK, there is representation of over eight parties in the House of Commons (HoC). Sartori moves to a description of 'relevant' parties as being valid. Following his definition, the UK and US have 'two-party' systems, while Germany has a 'moderate multi-party' system. This is on the basis of having between three and five 'relevant' parties. This would suggest the greatest similarity between the US and UK systems. However, the US (certainly at national level) is a complete two-party system, yet it is debatable whether the British Liberal Democratic party is truly 'irrelevant'. The inclusion of the Liberal Democrats would make the UK a moderate multi-party system, although, unlike in Germany, the third party still does not seem relevant in government. Further complication to these definitions is seen by post-war Germany always having either the CDU or SPD in power. This surely suggests some form of two-party domination. All of these complications must be noted when classifying party systems as the basis of analysis.

At non-national levels there are also some interesting trends to be noted. In the US, although party affiliation is often downplayed in elections at sub-national level, candidates will affiliate almost entirely with either the GOP or the Democrats. In the UK and Germany, local government is the scene of competition quite different to national level, with 'irrelevant parties' being seen to be relevant. In the UK the Liberal Democrats are the 'second party of local government' (displacing the Conservatives) and in various German Lander governments, there is representation of over five parties. This further complicated the classification above, showing that party systems must be seen at the different levels of politics.

Another important characteristic of a party system is the frequency of change between governments. Sartori argues, that the UK is more of a 'perfect' two-party system than the US. Alternation between the two parties is more frequent; "[in the US] the swing of the pendulum is not as it should be" . This suggests, as does Sundquist, that there are phases of single party domination in political history of the US, highlighted by the Democratic forty year domination of the House of Representatives. The German system best resembles the British system in the alternation of power between the major parties. The only possible post-war exception to this was the CDU-CSU and Tory domination throughout the 1980's and early 1990's.

It is also necessary to note the discipline within parties as a feature of the party systems. Again the US stands alone. As Professor Charles O Jones continually notes, the US is better described as a 'government of parties' rather than 'party government'. Whilst parties exist at all levels, obedience to a party line is at best limited. Presidents facing Congresses led by the opposing party are equally as productive as those facing Congresses of their own parties. Cross-party voting is far more common in the US than the UK or Germany, where the leadership is more powerful over party members. Equally, the President is unable to be removed from office by members of their own party, as the Prime Minister is in Britain, or the Chancellor is in Germany, suggesting the greater independence from party ties.

'Duverger's law' states that; "the simple-majority, single ballot system favours the two-party system". 'Duverger's hypothesis' states that; "the simple-majority system with a second ballot and proportional representation favours multi-partism". These observations seem to be true of all three of the countries above. Perhaps, the reason for two-party systems existing in the UK and the US are simply attributable to the single-member district elections. Such elections deter more than three parties competing as fragmentation of parties to more defined blocks, would allow the original opposition to gain power by splitting the vote. This is not the case in Germany, where a vote for any party has the possibility of national representation even if in a particular district there is a distinct minority. This would clearly explain the case of the US. In the UK, however, whilst in each constituency 'Duverger's law' seems to hold, the competition is not always between the two 'relevant' British parties, with the Liberal Democrats in contention in many areas. It is not inconceivable for the third party to win enough constituencies to prevent a clear majority of seats falling to a single party in the HoC. This cannot therefore be satisfactory outside each constituency so can't explain the national British party system.

The second theory important to numbers of parties is that suggested by both Sartori and Downs. They look at the distribution of different ideologies throughout the electorate. Where there is a simple 'bell-shaped' distribution, with a central peak of moderate ideology, there is a suggestion that parties will position themselves either side of the peak, competing, centripetaly for the moderate floating voters. When there is such a trend, there will be a limited number of parties. Sartori suggests either two-party systems or his category of 'moderate pluralist' systems will occur in such cases. Such observations do seem to match examples of the catch-all centrist parties of the UK, the USA and Germany. A contrast to such a model is seen in the examples of Israel and Italy, whose electoral ideology spectrum shows various peaks in its distribution. In such cases, parties form on or at either side of each peak, competing 'centrifugaly', away from the moderate centre ground.

Historically, there seems to have been little change within the party systems of these countries. The slight exception to this is the 1998 election in Germany. Previous to this, with the exception of the grand coalition of 1966-69, the FDP (liberal) party had been in perpetual power as junior coalition partner since the formation of the Republic. It is too early to tell if this displacement of the FDP as 'king-maker' will extend much further than the next election, so a view showing Germany's party system as vastly unchanged in its fifty year history is fair. The British and US party systems have seemed to be constant throughout the century and even in the 19th century the two party characteristics (albeit with different parties) was still in place. Such stability suggests either a lack of viable alternatives to the model, or else the entrenchment of the party system into the institutions of government. Such entrenchment would prevent other challengers from being able to upset the balance. This latter view is suggestive of Katz and Mair's model of 'cartel parties', which use the state apparatus to maintain their system between major parties and build barriers against new parties. Such claims of 'cartel parties' are certainly easy to level towards all three countries. The US makes gaining a place on the ballot hard (needing large nominations); Germany has a 5% electoral hurdle to gain powers and the UK shows refusal by the Conservative and Labour parties to be too sympathetic to Liberal calls for changes of the electoral system (although this stance is weakening under the current government).

The historical perspective also explains some of the differences in the frequency of changes of government between the parties. The US electorate is particularly fond of incumbents (averaging over 90% incumbent returns at federal level in this century). This means the frequency of changes is less than in more volatile electorates, as in the UK and Germany. Three or possibly four phases of single party domination seem apparent in the US during this century. Starting with Republican success from the civil war which was succeeded by the New Deal Democratic era which lasted through the presidency of Lyndon Johnson. After this, during the period of the Vietnam war and desegregation (unpopular in the traditionally Democratic south), the Republicans seemed to take the lead in Presidential races, with a Democratic Congress. The last phase involves a Democratic President and a house with a GOP majority. Such change is contrasted with the alternation of British governments and the seven different post-war German administrations. The forty year Democratic control of the House of Representatives, must show something about the stagnation of the US party system. Seemingly only major upheavals, such as the depression and race reforms have caused Sartori's 'pendulum' to swing in the United States.

The reason for the importance of history is largely to do with voter alignment and party identification. The US South, for example, is traditionally Democratic; the North of England traditionally Labour supporting and the North of Germany, traditionally supporting the CDU. Voters are often not willing to change support to any of the traditional opposition parties, so making the emergence of new parties hard to expect. The occasions that voters do change their votes are often as protests against their traditional party or in response to great social change. This is seen to occur more in the UK and Germany, where more centralised government can be more accountable for government actions than the US's 'government of parties' consensus.

The last approach is to look at the specific institutional and social features of the various countries which will account for their party systems. The features of government are perhaps the most obvious. In Britain, the single party in national government has a de facto monopoly of power, even accounting for the House of Lords. Other parties are relegated to often toothless opposition, usually through the media. The American, separated system allows for split-party government, forcing parties to negotiate and co-operate on specific issues, yet without formal coalition. This was seen by Clinton's success in passing legislating relating to NAFTA despite opposition from House Democrats. Unlike in Germany, the US executive and legislature are not completely dependent on one another and so the bond is ad hoc and fragile. The German system requires formal coalition building in order to achieve legislation. Such coalitions are strong and tests of them can cripple a government, such as the FDP's movement towards the CDU from the SPD before the 1980 election. These three styles of party governance have implications for the party systems that evolve to work within them. The increased party discipline in the UK and Germany is one consequence of this. Also, the infrequency of electoral shifts in the US results; the electorate is unable to know who to blame or praise for governing successes or errors.

The next point to raise is that of ideological cohesion within parties. In the US, whilst many groups typically identify with parties, there is absence of cohesion of ideology. For example, the Democratic party is seen as both the party of the Southern agricultural workers and the north-eastern urbanised liberals. This can result in the apparent contradiction that a Southern Democrat can be to the right on an ideological spectrum of a north-eastern Republican. This is undoubtedly a reason behind the lack of discipline within parties as well as the lack of grass-roots mass membership activity at national level. This is in contrast with the two European states, where the major parties have traditional ideological cleavages, around the political centre, yet maintaining core issues and backing. This has also led to a greater emphasis on party voting, compared to in the US where often incumbents campaign as if independents, rather than highlighting party policies.

This lack of ideological alignment in the US might be caused by the lack of a populist party of the working class. The Labour party and SPD filled this role in the UK and Germany. The absence of such a movement in the US could be due to success of a possible 'cartel' (from Katz and Mair's model); the lack of will due to relatively high standards of living or due to the practical problems of organising a new party in a country with fifty independent states and a population of over two hundred million. The result of the lack of workers movements is effectively electoral competition between two sets of management teams. This follows a Schumpeter like model of political competition.

The last observation to make, is that of the predicted decline of the party system. This is most associated with the US model, but has also be suggested of the UK and Germany. One of the key reasons for such predictions are those of candidate centred campaigning. This is becoming increasingly possible through modern media making mass membership party organisations unnecessary for campaigning. The lack of mass membership parties in the US means that this is perhaps easier to occur. Also, the absence of a need for a majority in the legislature to gain executive power allows for candidates to split from parties. The emergence of Ross Perot as a possible Presidential contender, without real party backing, backs this up. Such moves towards personality campaigning have been seen in the UK and Germany. The institutional composition of government, prevents the absence of parties in both these cases. The 'decline of the party system' when referring to these cases is therefore one of weakening of traditional voter alignment and movement towards more leader centred, media reliant catch-all campaigning.

In all, whilst there is a temptation to classify the three party systems are similar, the individual particularities make this classification largely unsatisfactory. There is some use of looking at factors such as the electoral system and ideological cohesion within each of the population as an explanation for the general trend of limited parties in all three states. Institutionally as well, it is clear that party systems will evolve in order to best suit their surroundings and then adapt to change these institutions to entrench themselves into it. For this reason it is hard to see what institutions lead to the evolution of such parties or what were formed as a result of the ways the parties, their ideology and their membership are. The decline of the party systems is also dependent on the institutions and their reliance on representative co-operation to work.