By Hans L. Zetterberg and Karin Busch

A remarkable stability of the party structure is one of the more noteworthy traits of the western democracies. A Swedish voter who in the 1920's fell asleep and woke up today would without difficulty recognize today's parties as the same as the ones operating half a century ago; the landscape of parties from the 1920's has not changed, although some party names would be new. By and large, the same is true for the parties on the continent: after the nazism and fascism and world war, the parties of the 1920's re-appear and gather the political ideas and feelings of the 1950's, 1960's and 1970's.

The changes of the social structure are slow and their political consequences are of little interest to the politician concerned about the next election. But, if we measure in decades instead of election campaigns, the structural changes become significant. For an overview of the changing bases of Swedish politics we shall compare more than 30.000 interviews from 1967-1969 with more than 23.000 interviews collected in 1946-1948. The older sample is a quota sample drawn by The Swedish Gallup Institute; the newer sample is a probability sample from The Swedish Institute of Public Opinion Research (SIFO).

At the time of the advent of Swedish democracy, there were large parts of the social structure that were not cultivated by any party. This may be illustrated by the fact that the turnout in election was low: in the 1910's between 47% and 70% of the enfranchised voted; in the 1920's, when women had obtained the right to vote, between 38% and 67% voted. In 1968, the turnout was 89%. About 94-96% now declare a party sympathy when pollsters ask their questions. There is no longer any important part of the social structure where political parties lack roots. It is this shortage of political no-man's-land that has made it difficult, almost impossible, for new parties to succeed, even if they happen to represent a large opinion. For example, about 2.3 million Swedes supported religious instruction in schools by signing a petition to the government. When a new party, The Christian Union, was later formed to push for this and similar causes it received less than 100.000 votes. Likewise, in 1968, strong sentiment among the public for a change in the organization of the non-socialist opposition resulted in insignificant voter strength for a new party, The Progressive Party, whose main platform was the merging of the non-socialist parties.

In the last few decades, Sweden has become more urbanized and more middle-class. The traditional party of the rural areas (Agrarian, Center) and the traditional party of the working-class (Social Democrats) have adjusted well to these changes in the social structure. This applies especially to the Center Party.

The advance of the Center Party into the cities is actually made at a faster pace than the immigration to the cities. A change of name in 1957, when the Agrarian Party started to call itself the Center Party, was a necessary part of this adjustment to an urbanized electorate.

In 1946 the Social Democrats already had a representation among white-collar workers amounting to 26% of the party, and this proportion has since increased so that 37% of the party are now outside the blue-collar working-class.

The breakthrough of the Social Democrats into the middle-class has kept pace with the structural decline in working-class occupations, and the expansion of the white-collar ranks. The so-called "pensions elections" at the end of the 1950's successfully consolidated the claim of Social Democracy to speak for all employees, both blue-collar and white-collar. The new middle-class territory was taken without loss among the traditional support in the working-class. In 1948, 71% of the working-class voted for the Social Democrats and in 1967-69, 68% did the same, an insignificant change.*)

The present anchorage of the politics to the social structure can be illustrated by two kinds of tabulations of occupations and parties. The first table shows the distribution of the party sympathies within different occupational categories. The second table shows the distribution of different occupations within different parties.

From this informative table we may conclude:

The political outlook of the white-collar employees depends to a considerable extent in what field they are employed.

On the whole, white-collar employees politically follow the owners of the enterprises in which they work. When the owners are Conservatives or Liberals — which mostly is the case within the private business sector — the employees become Conservatives or Liberals. If owners are mostly Centerites — as in farming and forestry — their employees support the same party. When the owners are Social Democrats — which mostly is perceived to be the case within the Government-owned business sector after nearly four decades of Socialist rule — the white-collar employees become Social Democrats. The party that argues for an increase of the public sector obtains most votes from there. By expanding the Government's role as owner of enterprises, the Swedish Social Democrats have increased their followers among administrators of these enterprises.

In the previous tables we saw how the occupational classes are divided into parties. Let us now turn the problem and study how the parties are combined according to occupation. We then get the following table:

The table indicates that many popular ideas about the social bases of the parties are incorrect or, at best, half-truths.

The class concept we use in social research separates occupational groups controlling larger and smaller resources, having responsibility for more and fewer employees, and requiring longer or shorter educational preparation. Differences in behavior, values, attitudes, knowledge, habits, life chances, etc. have been repeatedly documented between different social classes.

Different classes make claims on each other and on the old and new resources of the society. The political parties negotiate these claims, and attempt to formulate programs with special appeal to those classes within which they have or want to obtain votes. In this way they plead the causes of different classes of society, and it has been said that party politics is an organized, bloodless class war. In reality, as we have seen, all Swedish parties have sympathizers within all classes of society — albeit to rather different extents. The parties differ not only regarding the votes they receive from high and low class positions within society but also regarding the proportions of votes from central and peripheral parts of the social structure.

The centrality dimension divides people in the social structure according to their contacts. People with many contacts and information channels are central, those with few are peripheral. News and novelties usually spread from the central to the peripheral parts of the society. Within Swedish politics this dimension is interesting mainly because the central parts are usually favoured by the present economic developments and the peripheral parts are usually disfavoured, and this is true whether or not one is placed high or low on the class scale.

An interesting diagram is obtained when the political parties are positioned, so that it is possible to read their scores both in terms of centrality and class:

The circled mid-points of the crosses represent the median scores on the two scales. The length of the thin lines show within which region half a party's sympathizers fall, i.e., the terminal points of these lines lie at the 25th and 75th percentile.